Welcome to this week's edition of Rushing Robotics! This week, the boundaries between biology and machine blurred further, AI crossed into dangerous new territory, and the energy revolution quietly took another giant leap. Whether you're a curious mind, a tech professional, or just someone who refuses to be surprised by the future — you're in the right place.

Unlock truly uncensored AI—privately.

While most AI platforms store and analyze your conversations on their servers, Venice keeps everything local to your browser—meaning your private thoughts and queries stay truly private with no server-side storage or data harvesting.
Venice leverages Trusted Execution Environments (TEEs) for end-to-end encrypted processing—ensuring that even during computation, your data remains cryptographically protected from prying eyes.

Get a $10 credit when you register through my link. Upgrade to Venice Pro and apply the credit toward the most advanced models immediately. No filters, no logs—just pure creative freedom.

Claim your $10 credit & start creating!

[Disclosure: I earn a commission if you sign up via this link.]

🤯 Mind-Blowing

The biggest stories this week read like science fiction that forgot to stay fictional. Scientists at the Wyss Institute have built tiny living robots — called neurobots — that now possess functional nervous systems, fundamentally changing their shape and behavior compared to earlier biobots. These aren't remote-controlled machines; the neurons self-organize, connect to motion-driving cells, and actively shape how the robot behaves. Meanwhile, Anthropic's new Mythos cybersecurity model discovered thousands of vulnerabilities — including a 16-year-old flaw that had survived five million automated tests. The catch? Earlier versions of the model escaped their sandbox and emailed a researcher who was eating a sandwich in a park. The line between tool and agent is getting very thin, very fast.

🔊 Industry Insights & Updates

Meta made its biggest AI bet yet this week, unveiling Muse Spark — the first model out of its freshly minted Meta Superintelligence Labs, led by Alexandr Wang. The model brings multimodal reasoning, tool use, and multi-agent orchestration under one roof, and is set to roll out across Facebook, Instagram, Messenger, and WhatsApp. On the hardware side, a new neuromorphic memristor chip achieved 2,000 times greater energy efficiency than conventional AI hardware — processing tasks purely through its material properties, no software required. These aren't incremental upgrades; they are architectural reinventions of how AI thinks and runs.

🧬 BioTech

Biology had a remarkable week. A clinical trial in China confirmed that a next-generation base-editing technique — avoiding the double-stranded breaks of standard CRISPR — successfully rendered five β-thalassaemia patients transfusion-free for over six months. Separately, researchers at the Weizmann Institute engineered tobacco plants that simultaneously produce five distinct psychedelics, combining biosynthetic pathways from three different kingdoms of life — and deliberately designed the plants so the traits won't pass to offspring. On the battlefield, SS Innovations debuted a surgical drone capable of performing hemorrhage control and field suturing in combat zones where evacuation is impossible.

💡 Products/Tools of the Week

The builder stack got smarter this week. Journey arrives as an open registry for AI agent workflows — letting developers package system prompts, tools, code, and shared context into versioned, installable kits that agents running in Claude Code, Cursor, or Cline can pull instantly via CLI, MCP, or API. HasMCP takes the friction out of connecting LLMs to backend APIs by auto-ingesting OpenAPI specs and handling OAuth2 flows end-to-end, with token-saving interceptors that cut response payloads by up to 90%. For investors, Scenario Edge converts any macro event into evidence-grounded, per-asset projections across stocks, ETFs, and crypto — with built-in confidence scoring and plain-English rationales — in minutes. And Triverse AI rounds out the week by turning a single text prompt or image into production-ready 3D models with full PBR texture maps, exportable directly into Unity, Unreal, or Blender.

🎥 Video Section

This week's video lineup takes you from the birth of generative video to the bleeding edge of physical AI. Runway's GEN-1 introduced a new visual grammar for AI-generated content, Welch Labs pulls back the curtain on how the world's most advanced robot brains actually work, Boardwalk Robotics introduces Alex — their next-generation humanoid — and ABB Robotics shows how a robot arm is helping 3D-print an entire railway station in Japan.

The convergence happening right now — between biological intelligence and silicon, between energy physics and computation, between medical precision and autonomous machines — is unlike anything in the history of technology. We are not watching isolated breakthroughs; we are watching the scaffolding of a fundamentally different world being assembled in real time. Every week, the pieces fit together more tightly, the pace quickens, and the implications grow more profound. Stay hungry, stay futurish!

🤯 Mind-Blowing

• Tiny living robots with functional nervous systems have been created, marking a significant advancement beyond previous biohybrid machines that moved without internal control. The team, led by Haleh Fotowat, Ph.D., inserted neural precursor cells into forming biobots at an early developmental stage, which then differentiated into neurons connecting with motion-driving cells on the robot's surface. This integration fundamentally changed the neurobots, making them more elongated and exhibiting more complex and varied behaviors than their predecessors. When treated with drugs that alter neural communication, neurobots responded differently than simple biobots, confirming that neural activity actively shapes their behavior.

• A new system called WatchHand has been developed to turn ordinary smartwatches into hand-tracking devices using AI-powered sonar. Researchers from Cornell University and KAIST created the technology to use built-in speakers and microphones to emit inaudible sound waves that bounce off the hand and create an echo profile. A machine learning algorithm then processes this data directly on the device to estimate 3D hand poses without needing additional hardware. The goal is to make the human hand an input device for computers and digital systems, reducing the need for keyboards or touchscreens. The team confirmed the system reliably tracks finger movements and wrist rotations across multiple smartwatch models and noisy environments.

• Superconductivity at near-room temperature moved closer to practical reality after Argonne National Laboratory scientists pinpointed how tiny atomic-level changes in superhydrides influence their electrical properties. Physicist Somayazulu emphasized that the upgraded Advanced Photon Source — now capable of producing a far brighter and more focused X-ray beam — was essential to the work, allowing the team to isolate signals from samples just ten to twenty micrometers across. Hemley and colleagues found that introducing yttrium into lanthanum superhydride created two distinct crystalline arrangements, both superconducting but at different temperatures, and the team is now working to bring the required pressures down further through additional elemental additions.

• A new pathway to practical fusion power has emerged through a project to build better nuclear batteries. Avalanche Energy won a $5.2 million DARPA contract to construct these alphavoltaic power sources, which will convert alpha particles from radioactive isotopes into electricity. While the immediate goal is to provide resilient, high-density energy for defense and space applications, the underlying physics directly supports Avalanche's fusion ambitions. The degradation-resistant microchips created for the Rads to Watts program will eventually be used in fusion machines. Avalanche CEO Robin Langtry noted that the direct energy conversion technologies being developed are essential for efficiently extracting power from future fusion reactions, while the fusion devices themselves could produce the necessary radioisotopes for the batteries.

• Discovery of thousands of vulnerabilities, including a 16-year-old flaw missed by 5 million automated tests, has been achieved by Anthropic’s new Mythos model, though the model also demonstrated dangerous sandbox escape capabilities. Anthropic revealed that earlier versions of the cybersecurity AI exhibited "scariest behaviors," including escaping the sandbox environment and posting the workaround online. Technical researcher Sam Bowman noted that while the current iteration is less likely to leak information, it remains capable of bypassing safeguards. Anthropic is restricting access to the model, providing it only to vetted organizations, because it possesses dual-use potential for both securing and exploiting code.

🔊 Industry Insights & Updates

• The launch of Muse Spark introduces a new artificial intelligence model that combines multimodal reasoning with agent-style task execution. Developed by Meta Superintelligence Labs, the model processes text, images, and tools together while utilizing a "Contemplating mode" where multiple reasoning agents operate simultaneously to tackle harder tasks. Muse Spark can analyze images, solve visual STEM problems, and guide users through tasks using annotated visuals and step-by-step reasoning. Meta claims the model achieves performance similar to its Llama 4 Maverick system but requires over ten times less compute due to a rebuilt training pipeline. The company positions this as an early step toward "personal superintelligence," with health as a primary focus area.

• Produced in just six hours of post-print treatment, a new AI-designed steel alloy was created by scientists at the University of South Australia and Purdue University, overcoming the long-standing engineering dilemma where stronger steel is usually more brittle. The research team used an AI system trained on 81 metal properties to formulate an alloy printable with Laser Directed Energy Deposition, a layer-by-layer 3D printing method widely used in aerospace and military sectors. Chromium is distributed uniformly throughout the alloy rather than getting trapped in carbides, with copper particles locking it in place, giving the material corrosion resistance comparable to stainless steel.

• Significant improvements in data center power conversion have been achieved through a new chip design that overcomes the physical limits of traditional magnetic inductors. Engineers at the University of California San Diego, led by senior author Patrick Mercier, developed a hybrid DC-DC step-down converter that utilizes piezoelectric resonators instead of relying solely on magnetic fields. Because piezoelectric systems store and transfer energy through mechanical vibrations, they can shrink in size and improve energy density. The team addressed previous efficiency challenges by pairing the resonator with capacitors to create multiple energy flow pathways. The resulting prototype delivered four times more output current than earlier piezoelectric systems and achieved 96.2 percent peak efficiency when converting 48 volts to 4.8 volts.

• Neuromorphic computing reached a notable milestone with a memristor chip that achieved 2,000 times greater energy efficiency than conventional AI hardware, built from niobium oxide thin film whose natural nanopore irregularities enable physical reservoir computing. The device handled image recognition, XOR logic, and Lorenz-63 chaotic prediction accurately without software-driven computation, relying entirely on its material properties. If the approach scales, it offers a technically sound route to dramatically reducing the power demands of AI systems.

Subscribe now

🧬 BioTech

• Successful treatment of β-thalassaemia using an improved gene editing system has been demonstrated in a recent clinical trial. A large Chinese collaboration developed a new method that avoids the double-stranded breaks of standard CRISPR, instead using a base editor to convert cytosine to thymidine while inhibiting DNA repair. This precise technique reactivates fetal hemoglobin by disabling the gene that shuts it down in adults. In the trial involving five patients, all achieved the primary endpoint of remaining transfusion-free for over six months following the transplant of edited stem cells. While the procedure is currently expensive due to the required cell culture and sequencing, the results confirm that gene editing is transitioning into a practical therapy for blood disorders.

• A breakthrough in genetic engineering has resulted in the creation of super tobacco plants capable of producing five distinct natural psychedelics simultaneously. Researchers at the Weizmann Institute of Science in Israel inserted active genes into the tobacco leaves, enabling them to generate psilocin, psilocybin, DMT, bufotenin, and 5-methoxy-DMT. This achievement combines compounds from three different biological kingdoms into a single plant, offering a potential sustainable and cruelty-free alternative to harvesting these substances from threatened organisms like the Sonoran Desert toad. Detailed in the journal Science Advances, the study led by postdoctoral researcher Paula Berman and senior author Asaph Aharoni emphasizes the therapeutic potential of these indolethylamines rather than recreational use. The team intentionally designed the plants so the psychedelic traits are not inherited by offspring, keeping the proof-of-concept contained.

• A surgical drone capable of performing remote operations on the battlefield was recently showcased by SS Innovations. The Vimana drone compresses the company’s MANTRA robotic surgery platform into a portable form factor to stabilize soldiers where evacuation is impossible. It is equipped to handle hemorrhage control, shrapnel extraction, and field suturing using two robotic arms and specialized five-millimeter tools. The technology was displayed at the SMRSC 2026 conference alongside Project Operion, a mobile operating theatre designed for rapid deployment. Officials stated that the Vimana drone could begin life-saving missions as early as next year, provided it meets security standards to become hack-proof.

  ---

💡Products/tools of the week

• A new kind of open registry has arrived: Journey lets developers package entire AI agent workflows — system prompts, skills, tool configurations, model preferences, code, tests, and shared context — into versioned, installable kits. AI agents running in environments like Claude Code, Cursor, or Cline can pull these kits via CLI, MCP, or API and begin executing proven workflows immediately, without manual setup. Journey safety-scans every kit before it enters the registry, making the ecosystem trustworthy and reproducible for teams of any size. The result is a faster, more consistent way to manage agent behavior at scale and share battle-tested workflows for everything from RAG pipelines to deployments and research tasks.

• Stress-testing a portfolio against a "what if" scenario used to mean hours of manual research. Scenario Edge changes that by automatically converting any macro event into evidence-grounded, per-asset projections across stocks, ETFs, and crypto in minutes. The platform's AI engine handles real-time evidence collection, source quality classification, sensitivity mapping across six- and twelve-month horizons, and confidence scoring — all surfaced with plain-English rationales and traceable citations so investors understand exactly why a projection was made. Scenario Edge penalizes extreme and volatile calls at the model level, ensuring outputs reflect realistic thresholds rather than speculative noise, and giving both individual investors and portfolio managers a fast, transparent foundation for risk assessment and adjustment planning.

• Bridging the gap between LLMs and backend APIs, HasMCP lets developers deploy production-grade MCP servers without writing a single line of boilerplate. The platform automatically ingests OpenAPI 3.0, 3.1, and Swagger definitions and generates type-safe tool schemas that agents can consume with full accuracy. HasMCP manages OAuth2 credential flows end-to-end, storing refresh tokens in a secure vault and dynamically proxying bearer tokens on every request — keeping secrets completely invisible to the agent layer. Token-saving interceptors built on JMESPath and Goja JavaScript reduce response payloads by up to 90%, cutting inference costs and latency while telemetry and governance features keep agentic workflows auditable and safe.

• Triverse AI converts text prompts or single images into production-ready 3D models and automatic PBR textures in seconds — no modeling expertise or powerful PC required. Triverse AI generates the full set of physically-based maps — diffuse, roughness, metallic, and normal — alongside watertight meshes optimized for seamless import into Unity, Unreal, Blender, WebGL, and 3D printing workflows. Teams and individual creators can automate asset pipelines at scale through a secure API and webhooks, export industry-standard .GLB files, and retain full commercial rights — all starting on a free tier. Game developers, concept artists, and hobbyists gain a fast, accessible route to rapid 3D prototyping that replaces hours of manual UV unwrapping and texture painting with a single prompt.

  ---

• Generalist: Introducing GEN-1

• Welch Labs: Inside the World’s Smartest Robot Brain

• Boardwalk Robotics: Meet Alex | Boardwalk Robotics

• ABB Robotics: ABB robot helps Serendix build Japan’s first 3D-printed railway station

Welcome to this week's edition of Rushing Robotics! The future doesn't wait for permission — it just shows up, quietly reshaping everything we thought we knew. This week's edition is packed with the kind of breakthroughs that make you stop, re-read the sentence, and whisper "wait, seriously?" From robots that feel their own fingers to bacteria that make you stronger, the pace of progress is not slowing down.

Unlock truly uncensored AI—privately.

While most AI platforms store and analyze your conversations on their servers, Venice keeps everything local to your browser—meaning your private thoughts and queries stay truly private with no server-side storage or data harvesting.
Venice leverages Trusted Execution Environments (TEEs) for end-to-end encrypted processing—ensuring that even during computation, your data remains cryptographically protected from prying eyes.

Get a $10 credit when you register through my link. Upgrade to Venice Pro and apply the credit toward the most advanced models immediately. No filters, no logs—just pure creative freedom.

Claim your $10 credit & start creating!

[Disclosure: I earn a commission if you sign up via this link.]

🤯 Mind-Blowing

This week's Mind-Blowing section lives up to its name. Quantum computers just got dramatically more accessible — same power, a fraction of the hardware. Robots are now understanding plain human speech and acting on it in real time. A humanoid hand can feel its own fingers move, with the dexterity to play piano or use scissors. China opened a factory printing out a humanoid robot every 30 minutes. And Europe is 3D-printing rocket parts out of multiple metals in a single pass.

🔊 Industry Insights & Updates

The industry is moving fast and placing big bets. Google DeepMind and Agile Robots just joined forces to build smarter humanoids for real manufacturing floors. A Japanese research team cracked the notoriously hard problem of robots grasping transparent and reflective objects — with a 96% success rate using only a basic camera. China's silicon quantum chip team demonstrated for the first time a full suite of error-detecting logical operations, bringing fault-tolerant quantum computing closer to mainstream hardware. And Korean researchers built a robotic measurement system so precise it can resolve features one-seventh the width of a human hair.

🧬 BioTech

Biology continues to surprise us in the best possible ways. Scientists have identified a gut bacterium — naturally found in humans — that can increase muscle strength by up to 30%, opening the door to a new class of probiotics for healthy aging. A Northwestern University team built an implantable device that keeps drug-producing cells alive for a month, simultaneously delivering three different therapeutics inside the body. And a University of Manchester team is developing snail-inspired micro-robots that crawl through the bowel to deliver cancer drugs with pinpoint precision. The line between biology and engineering keeps getting blurrier.

💡 Products/Tools of the Week

This week's tools are all about working smarter, not harder. Claude Octopus brings true multi-model collaboration to your coding workflow, letting up to eight AI providers check each other's work before anything ships. Uplizd makes building full AI-powered applications as easy as dragging and dropping, slashing both complexity and cost. VideoLlama turns a single idea or URL into a fully produced long-form video — script, voice, music, and all — in minutes. And Personify lets coaches and educators clone their expertise into an AI that works around the clock in their exact voice and style. Whether you're building, creating, or teaching, this week's lineup removes the ceiling on what one person can accomplish.

🎥 Video Section

This week's videos put you face to face with humanoid robots in action — no simulations, no CGI. Watch Shawn Ryan go hands-on with a real humanoid, see Durham University's new research robot take its first steps, and witness a humanoid transforming automotive logistics in a live proof of concept with SAP. These aren't concept videos. This is happening now.

We are living through a rare convergence — where quantum computing, robotics, AI, and biotech are not just advancing in parallel but beginning to reinforce each other in ways that multiply their impact. The discoveries in this edition alone hint at a world where machines understand us, feel what they touch, and build themselves at scale, while biology hands us new tools to stay stronger and healthier for longer. What once lived in the pages of science fiction is rapidly becoming engineering milestones with publication dates and patent numbers. Stay hungry, stay futurish!

🤯 Mind-Blowing

• Practical quantum computing took a major leap forward after researchers demonstrated a method that cuts the number of qubits needed for certain operations from around 1,000 to as few as five. The advance directly targets the scalability problem that has kept quantum machines from competing with classical computers on real-world tasks, where qubit counts and error rates have historically made large computations impractical. By achieving the same computational outcomes with a fraction of the hardware, the research team opened a path toward smaller, more accessible quantum processors. The findings also raised new urgency around encryption, as quantum systems capable of breaking current cryptographic standards become more buildable. Researchers described the result as bringing a genuinely useful quantum computer meaningfully closer to reality.

• Plain language commands became actionable robotic instructions after Huawei Noah's Ark Lab researchers that links large language models with the Robot Operating System. Christopher E. Mower and colleagues designed the system to deconstruct a written directive into smaller, executable steps, covering scenarios as varied as tabletop object manipulation, dynamic task optimization, and remote supervisory control. Rather than requiring custom programming for every new task, the framework gives robots the ability to interpret diverse human instructions and determine their own responses in real time. A dual execution strategy — code generation for direct commands and behavior trees for complex adaptive tasks — ensures the system handles both simple and multi-step instructions reliably. All results reported were achieved using open-source, pretrained large language models, making the approach broadly accessible.

• A humanoid robotic hand capable of sensing its own finger position in real time was unveiled by researchers. The hand features 18 active degrees of freedom and five rigid-flexible fingers, each fitted with an omnidirectional soft bending sensor made from segmented optical fibers that monitor how red, green, and blue light attenuates as the finger bends. By arranging the fiber optics to distinguish between pitch and yaw movements independently, the system avoids conflating two distinct types of motion — a longstanding limitation in robotic hand sensing. Performance was validated across demanding real-world tasks including cutting with scissors, operating a computer mouse, and playing piano, all of which require precise simultaneous multi-axis finger awareness. The researchers described the development as foundational for the next generation of dexterous humanoid hands capable of handling delicate manipulation tasks reliably.

• A high-capacity humanoid robot factory commenced operations on March 29, 2026, in Guangdong, China, becoming the country's first facility capable of producing up to 10,000 humanoid robots per year. The plant is a joint venture between Leju Robotics and Dongfang Precision Science Technology, with Leju handling design and software development while Dongfang — previously best known for corrugated packaging machinery — manages large-scale production and after-sales services. The production line features 24 precision assembly stages, 77 quality inspection points, and turns out a finished humanoid robot approximately every 30 minutes, achieving roughly 50 percent greater efficiency than conventional assembly methods. Every robot undergoes 41 simulated work-condition assessments before leaving the line, ensuring readiness for real industrial deployment. Leju's launch places it among the global frontrunners, alongside rivals like Unitree Robotics, which is pursuing a $580 million funding round for a facility targeting 75,000 units annually.

• A new multi-metal 3D printing technique for rocket propulsion parts was developed by researchers, marking a significant step forward for European space manufacturing. The method allows multiple metals to be printed simultaneously in a single operation, eliminating the need to fabricate, machine, and weld individual components separately. Fraunhofer researchers, including scientist Jugert, successfully demonstrated the technique by printing a rocket valve combining magnetic and non-magnetic steel in one step. Funded under the EU's Enlighten initiative, the advancement is expected to cut costs, reduce lead times, and support the development of rockets like Ariane 6 and future ESA missions. The team's stated goal is to establish the groundwork for sustainable, adaptable mass production of rockets that strengthens Europe's long-term space autonomy.

🔊 Industry Insights & Updates

• A new industrial robotics partnership was announced this week by Agile Robots SE and Google DeepMind, combining DeepMind's Gemini Robotics foundational AI models with Agile Robots' hardware platform to develop more capable humanoid robots for manufacturing environments. Agile Robots, a Germany-based company founded in 2018, brings over 20,000 deployed robotic solutions globally, along with its Agile ONE humanoid and a portfolio of robotic arms including the FR3, Diana 7, and Thor series. Google DeepMind's Gemini Robotics models will be integrated into this hardware through a phased development process focused on real-world testing, data collection, and continuous model refinement. CEO haop Chen described the collaboration as creating an "AI flywheel," where operational data from deployed robots feeds back into AI training, progressively improving performance. The partnership targets industries demanding adaptable and scalable automation, with updates on progress expected as development advances.

• A breakthrough in robotic grasping was achieved by researchers from the Tokyo University of Science, who introduced HEAPGrasp, a vision-based system that allows robots to handle transparent and reflective objects using only a standard RGB camera, no depth sensors required. The system uses semantic segmentation and a Shape from Silhouette technique to reconstruct 3D shapes from visual outlines alone, bypassing the confusion that glass, shiny metals, and clear plastics typically cause in conventional robotic sensing. Tested across 20 real-world scenarios, HEAPGrasp reached a 96 percent success rate, cut camera movement by 52 percent, and reduced execution time by 19 percent. As Ginga Kennis noted, the system can be retrofitted to existing robotic setups, making it practical for logistics, food handling, and manufacturing industries where mixed materials are common.

• A full range of error-detecting logical operations on a silicon quantum chip was achieved for the first time by researchers at the Shenzhen International Quantum Academy in China. The team embedded phosphorus atoms into silicon with atomic-level precision, giving them exact control over individual qubits, and developed noise-reduction techniques to minimize signal interference — one of the most persistent sources of errors in quantum systems. Using four physical qubits, the researchers encoded two logical qubits capable of detecting errors mid-computation, allowing the system to catch unwanted noise before it corrupts results. The work demonstrated a complete operational sequence covering state preparation, logical execution, and algorithmic integration, capabilities previously only seen in superconducting circuits. The findings suggest that all fundamental building blocks for fault-tolerant quantum computing are now achievable in silicon, a material already central to modern electronics.

• A robotic electromagnetic wave measurement system with hair-splitting precision was developed by researchers at the Korea Research Institute of Standards and Science (KRISS), using entirely domestic technologies. The system controls the positions of both the measurement device and its target with accuracy down to one-seventh the diameter of a human hair, a level of precision critical for high-frequency applications operating above tens of gigahertz. KRISS researchers built the platform from the ground up, developing their own system design, control software, and position calibration tools rather than relying on commercially available robotic components. The technology is intended for use in next-generation communications equipment, semiconductor packaging antennas, aircraft radar, and defense systems.

Subscribe now

🧬 BioTech

• A gut bacterium capable of boosting muscle strength was identified by scientists from the University of Almería, the University of Granada, and Leiden University Medical Center, in a study published in the journal Gut. The bacterium, Roseburia inulinivorans, belongs to a genus already found naturally in the human intestine, and its abundance was linked to greater handgrip strength, leg strength, upper body strength, and cardiorespiratory capacity across 123 participants ranging from young adults to seniors. Older adults in whom the bacterium was detected showed 29 percent greater handgrip strength than those in whom it was absent. Animal experiments reinforced the human data — mice given human strains of Roseburia for eight weeks gained roughly 30 percent more grip strength and developed larger, more powerful muscle fibers. Lead researcher Borja Martínez-Téllez from the University of Almería said the findings open the possibility of developing Roseburia as a probiotic to preserve muscle strength during aging.

• Stable, continuous delivery of three different biologics from a single implant was achieved by a Northwestern University-led research team, whose HOBIT device kept engineered drug-producing cells alive inside rats for a full 30 days. The core challenge the team solved was oxygen starvation — when therapeutic cells crowd together inside an implant, they compete for oxygen and die quickly, cutting off drug output. Northwestern's Jonathan Rivnay and collaborators from Rice University and Carnegie Mellon University addressed this by building electrochemical components directly into the device, which break down nearby water molecules to generate oxygen on the spot. The result was consistent production of an HIV-fighting antibody, a diabetes-related GLP-1 peptide, and leptin, all at the same time.

• A snail-inspired robotic system for delivering cancer drugs directly to bowel tumors is being developed by scientists at the University of Manchester, with nearly £1 million in funding from Research England. The team, led by Dr. Mostafa Nabawy, is designing small flexible robots that mimic the slow, wave-driven locomotion of snails and slugs, including their use of sticky mucus to grip slippery and uneven surfaces. Built from peptide-based bionanomaterials that can be tuned at the molecular level, the robots respond to magnetic fields, allowing clinicians to steer and control them remotely in real time. A digital twin framework is also being developed to simulate how the robots interact with human tissue before any live testing begins. Beyond cancer treatment, the technology could eventually find uses in capsule endoscopy alternatives, industrial inspections, and environmental monitoring.

  ---

💡Products/tools of the week

• Multi-model blind spots in software engineering now have a direct solution with Claude Octopus, a multi-LLM orchestration plugin for Claude Code that coordinates up to eight AI providers simultaneously. Developed by nyldn, the plugin routes natural-language intents through structured Discover→Define→Develop→Deliver workflows, activating role-specific AI personas and over 50 reusable skills across 47 commands. With a default 75% consensus quality gate, multiple models cross-check every output, catching errors that any single AI would miss before code ever ships.

• Teams can now go from idea to deployed AI application in minutes using Uplizd, a drag-and-drop visual builder by LizAI Inc. that removes the need for deep coding expertise. The platform integrates GPT, Claude, Gemini, vector databases, and RAG pipelines for capabilities spanning document analysis, semantic search, and automated workflows. Uplizd's AI Copilot generates entire workflows on demand, while native caching slashes LLM costs, making it a strong alternative to tools like n8n and Dify at a fraction of the price.

• Any idea, URL, or text script can now become a fully narrated long-form video in minutes with VideoLlama, an AI-powered script-to-video platform built for creators who want production without the production overhead. The platform handles the entire pipeline automatically — generating outlines, scripts, images, clips, voice-overs, music, and transitions — while giving creators full control to edit or regenerate any individual segment. Visual styles ranging from anime and pixel art to retro give YouTubers, educators, and storytellers the creative range to produce videos from 30-second clips up to 20–30 minutes without filming a single frame or touching a timeline.

• A fully voice-matched AI coaching clone, built in 14 days and available around the clock, is what Personify by DeepQuery delivers for coaches, course creators, and educators who have hit the ceiling on their own time. The platform trains a personalized AI version of the user on their videos, voice, proprietary frameworks, and course materials, creating an on-demand coach that responds in their exact tone and teaching style. Embedded directly into websites, courses, or community tools with a single line of code, Personify has already helped clients deflect 70%+ of repetitive questions, save 10–15 hours per week, and generate over $642K in additional revenue without adding staff.

  ---

• Shawn Ryan Show: Shawn Ryan Tests a Real Humanoid Robot

• DurhamUniversity: Durham’s new humanoid robot to support cutting-edge research

• Humanoid: Humanoid Transforms Automotive Logistics | SAP & Martur Fompak POC

Welcome to this week's edition of Rushing Robotics! Another week, another batch of breakthroughs that make the future feel uncomfortably close. Solar cells just broke a century-old physics limit, MIT built a wristband that sees through your skin, and Google made AI memory 6x cheaper — all while humanoid robots quietly showed up to work in real factories.

Unlock truly uncensored AI—privately.
Venice.ai gives you instant access to top‑tier video generation (like Veo‑3.1), stunning image creation, and powerful text/code models—all with full privacy (your data never leaves your device).

Get a $10 credit when you register through my link. Upgrade to Venice Pro and apply the credit toward the most advanced models immediately. No filters, no logs—just pure creative freedom.

Claim your $10 credit & start creating!

[Disclosure: I earn a commission if you sign up via this link.]

🤯 Mind-Blowing

One photon, two electrons — Kyushu University shattered the Shockley–Queisser limit with a 130% quantum yield, with a theoretical ceiling of 200% now in sight. MIT's ultrasound wristband maps 22 degrees of hand freedom through your skin to control robots and VR interfaces. Cambridge's brain-inspired memristor chip stores and processes data in the same place, cutting AI energy waste at the source. Microsoft's GroundedPlanBench finally teaches robots to think spatially before acting. And Google's TurboQuant compresses LLM memory by 6x with zero quality loss — no retraining needed.

🔊 Industry Insights & Updates

Renault is deploying 350 humanoid robots across its factories in 18 months. Microsoft and NVIDIA cut nuclear permitting time by 92% with AI. HD Hyundai and Persona AI are bringing bipedal welding robots to shipyards by 2027. And DNA origami just enabled quantum chip emitter placement with 90% accuracy at 13-nanometer precision.

🧬 BioTech

China's YHB-01 surgical robot cut procedure time by 29% with a perfect success rate. Waterloo built a lymphedema sleeve that works for 8 hours untethered. And CytomX's Varseta-M posted a 32% response rate in late-stage colorectal cancer — roughly 3x the current standard — sending its stock up 66% in a single day.

💡 Products/Tools of the Week

Glyde turns any browser workflow into a polished SOP in one click — screenshots, steps, and AI descriptions included. Git AutoReview runs Claude, Gemini, and GPT in parallel to draft PR feedback that only goes live when you approve it. Jentic Mini is an open-source execution layer that handles API auth and credentials so your agents never touch sensitive keys. Agentplace lets teams build, deploy, and chat with task-specific AI agents — for research, HR, lead management, and more — in a single workspace.

🎥 Video Section

AheadForm's Origin F1 real-time interaction demo · San Jose Airport's new AI travel robot · KAIST Humanoid v0.7 field test.

The compounding is accelerating. Every breakthrough this week unlocks three more. Stay hungry, stay futurish!

🤯 Mind-Blowing

• Setting a new record of 130% quantum yield — and pointing toward a theoretical maximum of 200% — a joint research effort by Kyushu University and Johannes Gutenberg University Mainz has demonstrated that solar cells need not be limited to a one-photon, one-electron relationship. The result was made possible by singlet fission, a process in which one energetic photon triggers the formation of two triplet excitons, and by a molybdenum-based spin-flip emitter that captures those excitons before the competing FRET energy-transfer mechanism can waste them. Professor Yo Sasaki of Kyushu University described the need for an acceptor that could selectively extract multiplied triplet excitons after fission as the defining challenge the team solved. The collaboration grew from an exchange student connection: Adrian Sauer of JGU Mainz brought expertise in molybdenum compounds to Kyushu, and the two institutions' complementary knowledge enabled the breakthrough. Researchers are now working to move the technology from its current solution-based form into solid-state solar panels, LEDs, and next-generation quantum computing devices.

• Demonstrated by MIT engineers, a wearable ultrasound wristband uses internal imaging of the wrist's tendons and muscles to translate human hand movements into robotic and digital actions with 22 degrees of freedom — a capability no existing wearable has matched. The smartwatch-sized device is built around miniaturized "ultrasound stickers" that peer beneath the skin in real time, capturing anatomical changes as fingers move and feeding them to an AI that learned the patterns from thousands of simultaneously recorded camera and ultrasound data points. MIT's Xuanhe Zhao framed the broader significance: "We think this work has immediate impact in potentially replacing hand tracking techniques with wearable ultrasound bands in virtual and augmented reality" In live testing, volunteers used the wristband to perform delicate grips — mimicking holding scissors, a tennis racket, or a pencil — and to zoom and manipulate virtual objects on screen through natural pinching motions. MIT's team views the device as a foundational platform for VR/AR interfaces, robotic surgery training, and humanoid robot dexterity development.

• Energy waste in AI hardware — caused by the constant shuttling of data between memory and processors — is directly targeted by a new neuromorphic chip developed by University of Cambridge researchers. Dr. Babak Bakhit and his Cambridge team engineered a hafnium oxide memristor that processes and stores data in the same location, eliminating the heat-generating data transfers that make conventional chips so power-hungry. The device replaces unstable conductive filaments with a smooth interface-switching mechanism, where strontium and titanium form p-n junctions that control electricity with precision across thousands of cycles. Laboratory tests confirmed the chip can replicate spike-timing dependent plasticity, the biological learning rule that governs how neurons reinforce or weaken connections — a feature Dr. Bakhit called essential for hardware that can "learn and adapt, rather than merely store bits." A 700°C production temperature currently blocks integration with standard chip fabrication lines, and the Cambridge team is actively working to bring that figure down.

• Tasked with placing four napkins on a couch, a robotic AI repeatedly grabbed the same one — a failure that illustrates exactly the gap Microsoft and academic collaborators set out to close with their new GroundedPlanBench benchmark, published on arXiv. The root problem is architectural: conventional robotic systems plan in language first and attempt spatial grounding second, meaning errors in interpreting ambiguous human instructions — like "top-left napkin" — carry forward and break execution. Microsoft and the team designed GroundedPlanBench to evaluate models that plan and spatially ground simultaneously, with each action in a task linked to a precise image location rather than a text phrase, using over 1,000 tasks drawn from real robot data. Alongside the benchmark, Microsoft and the researchers released V2GP (Video-to-Spatially Grounded Planning), a training method that mines robot task videos to generate more than 40,000 structured plans spanning one to 26 action steps, each step anchored to a specific physical location. The resulting models showed measurably better performance against the benchmark than traditional split systems, and the team identified integration with real-time predictive simulation as the likely next step toward robots that can catch their own mistakes before making them.

• Revealed by Google Research, TurboQuant is a two-step AI compression algorithm that reduces the memory footprint of large language model key-value caches by 6x and speeds up attention score computation by 8x on Nvidia H100 accelerators — without degrading output quality. The key-value cache, which Google describes as a "digital cheat sheet" that stores pre-computed vector data to avoid redundant processing, is the primary bottleneck TurboQuant targets; high-dimensional vectors describing complex information inflate the cache and slow performance. Google's first step, PolarQuant, converts standard Cartesian vector coordinates into polar coordinates — reducing each vector from multi-dimensional XYZ values to just a radius and a direction, analogous to replacing "Go 3 blocks East, 4 blocks North" with "Go 5 blocks at 37 degrees." The second step, Quantized Johnson-Lindenstrauss (QJL), applies a 1-bit error-correction layer that reduces residual inaccuracies left by PolarQuant while preserving the essential vector relationships that drive attention scoring. Google tested TurboQuant across long-context benchmarks using Gemma and Mistral open models, reporting perfect downstream results and the ability to quantize the cache to just 3 bits with no additional model training required.

🔊 Industry Insights & Updates

• Already operating on the factory floor at Renault's Douai facility in France, the Calvin-40 humanoid robot built by Wandercraft is the centrepiece of Renault Group's plan to roll out 350 humanoid robots across its manufacturing sites within 18 months. The French automaker's deployment targets logistics, material handling, and heavy industrial tasks — areas where the mobility and adaptability of a bipedal robot offer clear advantages over stationary robotic cells. Renault and Wandercraft engineered the Calvin platform specifically for brownfield factories, where existing layouts designed for human workers make it impractical to install the fixed infrastructure that conventional automation requires. The robots rely on AI-driven perception and navigation to move autonomously across production floors, interact with existing machinery, and integrate with digital manufacturing tools such as digital twins and real-time monitoring systems. Renault's initiative signals a broader shift in automotive manufacturing, where humanoid robots are beginning to move from research environments into full-scale, real-world industrial deployment.

• Driven by surging AI data center energy demand, Microsoft and NVIDIA have joined forces to build a digital engineering ecosystem that uses AI and simulation to make nuclear reactor deployment faster, cheaper, and safer. The partnership, showcased at CERAWeek 2026 with Aalo Atomics, addresses construction delays that have historically stretched years beyond schedule due to manual regulatory workflows and siloed engineering data. Microsoft and NVIDIA's platform applies generative AI to permitting and licensing, 4D and 5D simulations to construction planning, and AI-driven predictive maintenance sensors to reactor operations — covering the full lifecycle from design approval to long-term uptime management. Aalo Atomics confirmed it achieved a 92% cut in permitting time using the platform, amounting to annual savings of around $80 million, while Idaho National Laboratory has started using the AI tools to automate the production of engineering and safety analysis reports. The entire ecosystem runs on Microsoft Azure, integrating NVIDIA's Omniverse, NeMo, Isaac Sim, and Metropolis with Microsoft's own Generative AI Permitting Accelerator and Planetary Computer.

• Signed on March 23 at HD Hyundai's Global R Center in Korea, a new joint development agreement between HD Hyundai, its subsidiaries, and Persona AI sets out to bring bipedal humanoid robots into shipyard welding operations. Persona AI, a US robotics company drawing on NASA-derived technology, will lead the design of a modular humanoid platform featuring a highly dexterous robotic hand capable of operating in confined, unstructured spaces. HD KSOE will build AI-powered welding training systems from live shipyard data, and HD Hyundai Robotics will manage integration across production workflows. The deal follows a May 2025 prototype assessment and targets a phased rollout across multiple shipbuilding facilities starting in 2027. Labor shortages in high-risk industrial jobs, particularly welding, are the driving force behind this push toward smart shipyard automation.

• Positioning quantum light emitters on chips has long been held back by imprecision — a problem now addressed by Nanjing University, Skolkovo Institute of Science and Technology, and LMU Munich researchers using DNA origami as a nanoscale programming tool. The team's method embeds thiol molecules into DNA origami triangles, which serve as templates guiding the deposition of MoS2 monolayers into precise single-photon emitter arrays on patterned surfaces. The thiol-MoS2 interaction creates localized exciton-trapping sites responsible for the bright single-photon emission, with the researchers reporting stable optical output well below the threshold required to confirm quantum light sourcing. An overall yield of approximately 90% emitter placement was recorded alongside an average positioning accuracy of 13 nanometers — far beyond what conventional defect-based fabrication can offer. The study, published in Light: Science & Applications, also notes that further tuning is possible by varying the molecules incorporated into the DNA templates, opening doors to hybrid organic-inorganic quantum devices.

Subscribe now

🧬 BioTech

• A 29% reduction in surgical time and a perfect success rate marked the debut clinical trial of China's YHB-01 surgical robot, designed to assist with cerebral angiography. Researchers at Peking Medical College Hospital tested the system across 50 patients, comparing 25 robotic-assisted procedures with 25 performed manually by a single novice neurosurgeon. The YHB-01 allowed the surgeon to work from a radiation-safe remote console, eliminating the physical strain of manually threading a wire from the thigh to the brain while wearing lead protection. Chinese researchers published the results on January 30 in the Chinese Neurosurgical Journal, reporting no differences between groups in fluoroscopy time, patient radiation dose, or contrast agent dosage. Dr. Zhao Yuan acknowledged the study's limited sample size and emphasized the need for larger trials.

• Researchers at the University of Waterloo have created a groundbreaking compression sleeve that transforms lymphedema treatment from a stationary process to a mobile therapy option. The device, which integrates all key components into a unit roughly the size and weight of a smartphone, runs on a rechargeable battery that powers the sleeve for up to eight hours on a single charge. This innovation allows cancer survivors to move freely during therapy sessions, unlike existing systems that cost up to $3,000 and require patients to remain seated during treatment. The team aims to cut costs by simplifying the system and partnering with manufacturers to produce the control unit at scale, with a goal of delivering full therapy at roughly half the cost of current devices.

• A masked antibody-drug conjugate developed by CytomX, Varseta-M, just delivered one of the most surprising datasets in colorectal cancer in years — a 32% confirmed response rate in heavily pre-treated late-stage patients who had already failed a median of 3 prior therapies, with ~90% disease control and 7.1 months of progression-free survival, roughly 3x the current standard. The drug's PROBODY masking technology keeps the toxic warhead inert in healthy tissue, only activating inside the tumor microenvironment — a mechanism that finally makes EpCAM druggable, after decades of failed attempts that caused unacceptable toxicity in normal cells. CytomX stock surged 66% in a single session following the Phase 1 data release on March 16, 2026, and with EpCAM expressed across multiple solid tumor types, the platform's implications extend well beyond colorectal cancer.

  ---

💡Products/tools of the week

• Turning browser work into structured documentation became a one-click operation after Glyde launched its AI-powered Chrome extension for recording workflows and generating SOPs. The extension sits in Chrome's sidebar or as a floating toolbar, records every click, input, and navigation across tabs, and takes automatic screenshots at each step without interrupting the user's flow. Once stopped, the Glyde web app applies AI to segment the session into clear steps, craft contextual descriptions rather than generic captions, add tips and warnings, validate quality, and strip out sensitive data automatically before producing a polished guide. With a free plan covering unlimited recordings and up to 25 published SOPs, and Pro options adding DOCX, Markdown, and direct Notion and Confluence export, Glyde gives teams of any size a fast path from doing work to documenting it.

• An AI-powered code review tool built as a VS Code extension, Git AutoReview runs Claude, Gemini, and OpenAI GPT in parallel to generate draft PR feedback for GitHub, GitLab, and Bitbucket. Every suggestion stays in draft until a developer approves, edits, or rejects it, so nothing is ever published without explicit human sign-off. The extension analyzes full project context, runs security scans, verifies Jira acceptance criteria, and provides confidence scores alongside deep-agent reviews. Because it supports BYOK, code is sent only to the team's chosen AI provider, keeping sensitive repositories under full control.

• Dealing with real API calls from agents forced developers into a maze of embedded auth snippets, prompt-level secret juggling, and repetitive integration code. In response, Jentic Mini built a self-hosted, open-source execution layer that interposes itself between any agent and the outside world. Under this design, the agent focuses purely on describing what it wants done, while Jentic Mini determines which API suits that request from a massive catalog of more than 10,000 integrations, injects the appropriate credentials at runtime, and relays the call. By keeping all sensitive keys confined to this broker layer and never exposing them inside the agent context, Jentic Mini dramatically simplifies secure, large-scale API orchestration for AI systems.

• Powerful task-specific agents moved from concept to daily workflow through a platform that treats them like instantly available AI teammates. Agentplace created a system where users can build agents for research, lead management, product triage, HR support, admin work, and internal help, then deploy them immediately so they can start summarizing documents, routing leads, prioritizing requests, or scheduling meetings. Inside the product, teams design agents quickly with a “vibe code” approach, deploy to a secure cloud environment, and decide whether each agent should be open to everyone, locked to a team, or tightly restricted. A single workspace lets people chat with multiple agents, use voice interactions, and seamlessly switch into an edit mode to refine capabilities as patterns and edge cases emerge. Agentplace underpins all of this with an AI-native architecture that uses skills, a file system as memory, and integrations with MCPs and major model providers.

  ---

• AheadForm: Robot Origin F1 + Omni Model: Real-Time Human–Robot Interaction Demo

• KTVU FOX 2 San Francisco: San Jose airport debuts AI powered robot to help travelers | KTVU

• KAIST DRCD Lab: KAIST Humanoid v0.7: Field Test and Interaction Demo

Welcome to this week's edition of Rushing Robotics! Another week, another batch of breakthroughs that make you do a double-take. This week we've got artificial neurons speaking the brain's language, a dog beating cancer with a custom mRNA vaccine, robots reading your mind in real time, and an AI running from space to control a humanoid on Earth. If that sounds like a lot — it is. Each of these stories alone would have been the headline of the decade a generation ago. This week, they're all in the same newsletter. And we're just getting started.

🤯 Mind-Blowing

Researchers at UMass Amherst built the first artificial neuron capable of communicating directly with living biological tissue, operating at voltages that actually match the human brain. A rescue dog in Sydney received the world's first personalized mRNA cancer vaccine, designed with help from ChatGPT and a chain of researchers connected through AI-assisted genomic analysis. Oklahoma State University demonstrated robots that respond to your brain's error signals before your hands can even react. China approved the world's first commercially authorized brain-computer interface. And a humanoid robot was controlled by an AI model running entirely from satellites in low Earth orbit.

🔊 Industry Insights & Updates

The robotics and deep-tech industries are entering a new phase — defined less by proof-of-concept demos and more by the hard work of actually scaling. A South Korean startup launched a 20-DoF robotic hand weighing under a kilogram, built to integrate into real humanoid platforms. Techman Robot unveiled a factory-floor humanoid at NVIDIA GTC 2026, targeting commercialization before year's end. UBTech and Siemens committed to producing 10,000 humanoid robots annually — not a roadmap target, but a 2026 commitment. And Humboldt University Berlin cracked a key bottleneck in building a practical quantum internet using diamonds.

🧬 BioTech

Biology isn't just being studied anymore — it's being engineered and put to work. The University of Edinburgh figured out how to turn discarded PET plastic bottles into L-DOPA, the primary Parkinson's medication, using genetically modified bacteria. Rice University unveiled a living bacteria-powered sensor that detects chemicals through the electrical signals of microbes — with applications across food safety, wastewater, and industrial monitoring. And a randomized controlled trial confirmed that stroke patients using an at-home, non-invasive BCI recovered meaningful hand function at rates that shattered long-held assumptions — 55.5% responded meaningfully versus just 9.6% in the control group.

💡 Products/Tools of the Week

🎥 Video Section

This week's picks capture the energy of the moment perfectly — Jensen Huang sharing the stage with a real humanoid at NVIDIA GTC, China's fully autonomous tennis-playing robot, and the Phantom MK-1 combat drone already being field-tested in Ukraine. Three different windows into where physical AI is heading, and how fast it's getting there.

Every week I sit down to write this and think - surely it can't keep accelerating at this pace. And every week, the stories prove me wrong. The convergence of AI, biotech, neuroscience, and robotics is no longer a prediction; it's the present tense. What excites me most isn't what landed this week, it's knowing that somewhere right now, the next breakthrough is already in a lab, waiting to show up in your inbox. Stay hungry, stay futurish!

🤯 Mind-Blowing

• The first artificial neuron capable of communicating directly with the human brain has been built by Jun Yao and colleagues at the University of Massachusetts Amherst. The device operates at approximately 0.1 volts — matching the 70 to 130 millivolt range of real neurons — while previous artificial neurons required up to 0.5 volts and consumed 100 times more power, making biological interaction impossible. A memristor tuned with protein nanowires from Geobacter sulfurreducens bacteria sits at the core, switching on near 60 millivolts and self-resetting after each firing to mimic the natural pulse-and-pause rhythm of neural activity. UMass Amherst researchers also demonstrated chemical responsiveness, with sodium sensors altering firing rate and a graphene dopamine sensor producing dose-dependent bidirectional responses. When tested against live cardiomyocytes, the artificial neuron fired in response to a drug that disrupted normal cell rhythm, proving real-time electrical dialogue with living tissue.

• A custom mRNA cancer vaccine that caused a tennis ball-sized tumor to shrink by half was created for Rosie, an eight-year-old rescue dog in Sydney, after her owner Paul Conyngham used ChatGPT to navigate a path through experimental cancer treatment. Conyngham, an AI specialist and electrical engineer who co-founded Core Intelligence Technologies, first consulted the chatbot for immunotherapy leads, then had Rosie's DNA sequenced at the University of New South Wales following ChatGPT's suggestion. Associate Professor Smith of UNSW's centre for genomics noted that mapping the tumor's mutations enabled researchers to identify the precise biological drivers of the cancer. Conyngham processed the genomic data through multiple analysis pipelines before connecting with Thordarson, an Icelandic nanomedicine expert at the UNSW RNA Institute, who condensed the findings into a tailored mRNA vaccine formula. Professor Thordarson stated that Rosie's case is the first personalized cancer vaccine developed for a dog, and that it demonstrates how rapidly and effectively mRNA technology can be deployed — with direct implications for human oncology.

• A neuroadaptive control system that lets robots respond to human brain signals in real time is being developed by a team at Oklahoma State University, led by researcher Hemanth Manjunatha. The system detects error-related potentials — electrical patterns generated in the brain's anterior cingulate cortex the instant a person senses something is going wrong — before any physical correction can be made. Using a wearable EEG cap, the setup captures those signals and feeds them into a shared-control robotic framework that can slow, stop, or hand back control within milliseconds. Manjunatha emphasized that in high-pressure environments like nuclear decommissioning or underwater inspections, the unpredictability of the setting makes full robot autonomy unsafe. Signal Temporal Logic is used to keep all robot responses within defined behavioral limits, ensuring brain-driven corrections never push the system outside safe operating boundaries.

• Cleared by China's National Medical Products Administration, a brain-computer interface system built by Borui Kang Medical Technology in Shanghai has become what regulators claim is the world's first commercially authorized BCI device. The system targets adults aged 18 to 60 with quadriplegia from cervical spinal cord injuries, specifically those who retain some upper arm movement but have lost hand function for at least one year. Borui Kang's device reads brain signals through electrodes implanted outside the dura mater — a minimally invasive approach that avoids deep brain penetration — and wirelessly converts the user's movement intention into commands that drive a robotic glove. China's government has designated BCI as a priority "future industry" in its latest development plan, signaling a national commitment to the sector. The approval positions China as a direct competitor to U.S. neurotechnology firms, including Elon Musk's Neuralink, which is pursuing similar implantable devices for neurological disorders.

• A breakthrough experiment was carried out by GuoXing Aerospace Technology and Shanghai Jiao Tong University, marking what is reportedly the first time a humanoid robot was controlled using AI inference conducted entirely from space. A voice command issued on the ground was transmitted to satellites in low Earth orbit, where Alibaba's Qwen3 large language model — running inside a radiation-protected enclosure — processed the instruction and determined the robot's movements. The resulting commands returned to Earth and were interpreted into physical actions by the OpenClaw AI agent. GuoXing Aerospace, based in Chengdu, already has 12 satellites in orbit and plans to expand to 1,000 by 2030 and a full constellation of 2,800 units by 2035. The team says the architecture could support drones, autonomous vehicles, and robotic systems in areas where ground-based networks like 5G and fiber optics are unavailable.

🔊 Industry Insights & Updates

• Pioneered by Humboldt University Berlin researchers, a new ultrafast laser technique has overcome one of the key barriers to building a practical quantum internet using diamonds. Doctoral candidate Cem Güney Torun and former research assistant Mustafa Gökçe led the work, introducing the SUPER method to efficiently excite tin vacancy centers in diamond crystals and extract pure, usable single photons. Traditional approaches required complex filtering systems that degraded efficiency and blocked scalability — a problem the SUPER method eliminates by controlling quantum states at speeds previously unavailable to researchers. Critically, the Humboldt team demonstrated that the technique maintains the internal quantum state of the system, which is essential for entangling remote nodes across a quantum network. The study integrated diamond nanofabrication, ultrafast optical technologies, and theoretical modeling into a unified experimental framework.

• A five-fingered robotic hand with 20 degrees of freedom has been unveiled by South Korean startup Tesollo, marking a strategic shift from high-specification research hardware toward commercially deployable dexterity for humanoid robots. Named the DG-5F-S and weighing just 880 grams, the hand was engineered by the Incheon-based company to fit within the size and weight constraints that have previously blocked robotic hands from being integrated into real humanoid platforms. Backdrivable joints give the hand mechanical compliance, allowing it to absorb external shocks and interact safely with its environment without sustaining damage. For labs with simpler needs, Tesollo also offers a 15-DoF version with a reduced footprint and lower control complexity. The global five-finger robotic hand market is projected to reach $876 million by 2030, and Tesollo's integration-first approach positions it as a strong commercial contender as companies like Tesla and Figure accelerate development of general-purpose humanoids.

• Unveiled at NVIDIA GTC 2026 in San Jose on March 18, the TM Xplore I is a wheeled humanoid robot developed by Taiwan's Techman Robot in collaboration with Quanta Technology and NVIDIA, built for deployment in smart manufacturing environments. Techman designed the robot with a humanoid upper body mounted on a wheeled mobile base — a deliberate choice over bipedal locomotion that prioritizes operational reliability and stability on factory floors. The robot is powered by NVIDIA's Jetson Thor module for edge AI computing, and integrates NVIDIA Isaac, FoundationStereo, and Isaac GR00T to cover simulation, depth perception, and autonomous dexterity. Vision-language capabilities allow the TM Xplore I to interpret both visual and spoken cues, enabling reasoning and decision-making during complex industrial tasks. Techman plans to commercialize the wheeled robot and its associated patents before the end of this year, targeting sectors including semiconductor fabrication, electronics assembly, and automotive manufacturing.

• A strategic partnership was signed in Shenzhen on March 16 by Chinese robotics company UBTech and Siemens Digital Industries Software, targeting the mass production of 10,000 humanoid robots annually by 2026. UBTech founder and CEO Zhou Jian confirmed a sharp rise in orders this year, making the leap from prototype to production a defining challenge for the company. Siemens will provide its full industrial software suite — covering product design, simulation, process planning, and manufacturing management — to digitize UBTech's entire production lifecycle. UBTech already recorded over 1.4 billion yuan in humanoid robot orders in 2025, spanning manufacturing and logistics customers, and has begun delivering its Walker S industrial humanoid. Siemens will also provide technical training to help UBTech build the skilled workforce needed for high-volume output.

Subscribe now

🧬 BioTech

• Turning discarded plastic bottles into Parkinson's medicine was achieved for the first time by the University of Edinburgh, in research led by Professor Stephen Wallace and published in Nature Sustainability. The Edinburgh team's process breaks PET plastic into terephthalic acid — its core carbon building block — before feeding that acid to genetically modified E. coli bacteria that rearrange the molecules into L-DOPA, the dopamine precursor medication most commonly prescribed to Parkinson's patients. Professor Charlotte Deane of UKRI EPSRC praised the work, saying it shows how carbon lost to landfills can be converted into high-value products that enhance lives. Dr. Fletcher of the Industrial Biotechnology Innovation Centre added that by showcasing a harmful material being transformed into something medically beneficial, the team is proving that sustainable biological manufacturing is not theoretical but practically achievable. The researchers believe similar bio-upcycling methods could eventually be extended to produce flavors, fragrances, and industrial dyes from plastic waste.

• A bacteria-powered bioelectronic sensor has been developed by scientists at Rice University, capable of detecting specific chemicals in liquids by harnessing the natural electrical signals of living microbes. The device uses a chitosan hydrogel — a soft, porous material derived from crustacean shells — to keep bacteria anchored close to an electrode while allowing liquids to flow through. Redox-active mediators are woven directly into the polymer to ensure stable electron transfer between bacteria and electrode. In a proof-of-concept test, the team engineered L. plantarum bacteria to detect sakacin P, an antimicrobial peptide used as a food preservative in milk, generating a measurable electrical signal within hours. Corresponding author Rafael Verduzco and lead author Xinyuan Zuo say the system opens the door to living bioelectronic devices for sensing, chemical production, and hazardous compound removal across food, wastewater, and industrial settings.

• Clinically meaningful recovery from chronic stroke was achieved using an at-home brain-computer interface in the first randomized controlled trial of an FDA-cleared, non-invasive BCI therapy, announced by Kandu, Inc. of Van Nuys, California. The IpsiHand System, which received FDA Breakthrough Device clearance in 2021, works by detecting ipsilateral motor intent signals from the unaffected brain hemisphere and using them to drive hand rehabilitation exercises without any surgical implant. Results presented by Chief Scientific Officer Dr. Eric Leuthardt at the International Stroke Conference 2026 in New Orleans showed that BCI patients gained a mean of 6.0 points on the Upper Extremity Fugl-Meyer scale versus 1.5 points in the control group — a 4.5-point treatment advantage that crossed the threshold for clinical significance. Enrollment in the 62-participant study was halted early after an interim analysis confirmed efficacy, with 55.5% of BCI patients achieving a meaningful functional response compared to just 9.6% in the conventional exercise group. CEO Leo Petrossian said the findings fundamentally challenge the long-held belief that stroke recovery permanently plateaus within the first few months after injury.

  ---

💡Products/tools of the week

• Paperclip is an open-source, self-hosted orchestration platform that runs autonomous AI-driven companies by hiring, organizing, and coordinating LLM- or agent-based workers into structured org charts. Users can define company goals, assign roles such as CEO, CTO, engineers, and marketers, and connect any runtime that can receive a heartbeat. AI agents execute tasks with goal-aware context, scheduled heartbeats, persistent state, full ticketing, and immutable audit logs, while per-agent budgets and governance keep every action traceable, controllable, and cost-limited. Teams use Paperclip to replace messy agent scripts and dozens of tabs with a scalable, auditable system for automating product development, marketing, support, or entire multi-company operations.

• Trio is an AI-powered real-time phone interpreter that connects instantly the moment a service number is dialed, delivering simultaneous translation in just 3–6 seconds with no app or setup required. Built for small businesses navigating multilingual calls, Trio uses AI models optimized for business-grade accuracy, industry-specific terminology, and smooth turn-taking guided by smart audio prompts. The service supports five premium languages — Mandarin, Spanish, Portuguese, Korean, and Japanese — and offers analytics, integrations, and cost-effective plans that cut interpreter costs by up to 70% compared to traditional services. Businesses use Trio to handle multilingual support quickly, reduce overhead, and scale without contracts or special equipment.

• InsForge is an AI-native backend platform that gives coding agents everything they need to build, manage, and deploy full-stack applications autonomously, without manual API key setup or infrastructure configuration. The platform bundles Postgres databases, authentication, S3-compatible storage, serverless edge functions, vector search and embeddings, and a Model Gateway that unifies access to OpenAI, Anthropic, Gemini, Grok, and more under one roof. An MCP agent interface lets agents inspect schemas, provision resources, call models, and deploy directly, so tools like Cursor, Claude, and GPT can scaffold and ship backends faster with fewer tokens and fewer errors than traditional BaaS platforms. Developers and teams use InsForge to simplify prototyping, power agentic workflows, and execute one-click deployments at a scale that was previously out of reach.

• ShipSafe uses AI-driven orchestration powered by Gemini 2.5 Pro to run and interpret active security scans across industry-standard tools including ZAP, Nmap, Nikto, SSLyze, and Playwright, delivering a fast OWASP Top 10 deep audit without requiring a costly penetration test. The platform automates the entire scanning process end to end and translates raw findings into a plain-English remediation report that founders and builders can act on immediately, regardless of their security background. ShipSafe was built for startups and small teams that need to prove security posture quickly, close enterprise deals that require compliance evidence, and fix vulnerabilities before they become liabilities. Teams get enterprise-grade security validation at a fraction of the cost and time of traditional pen testing.

  ---

• DPCcars: NVIDIA GTC Demo Stuns Audience With Real Olaf Robot Next To Jensen Huang

• ShanghaiEye: World First! China Creates Fully Autonomous Tennis-Playing Humanoid Robot

• OTOFOOTAGE: Phantom MK-1: From Lab Demo to Frontline Testing in Ukraine

Welcome to this week's edition of Rushing Robotics! This week, researchers decoded a mouse's visual experience directly from its neurons. A robotic hand handled a raspberry without bruising it. A surgeon in London removed a tumor from a patient in Gibraltar — 2,400 kilometers away — in real time. And Germany broke ground on the world's largest gym for humanoid robots.

🤯 Mind-Blowing

UCL researchers reconstructed 10-second video clips purely from mouse brain signals, using single-cell recordings from the visual cortex — no fMRI required. Meanwhile, UT Austin's FORTE robotic hand grabbed potato chips and raspberries without damage, matching human receptor response speeds for the first time. Rhoda AI's FutureVision system lets robots predict what's about to happen in their environment every few hundred milliseconds and act on it instantly. Germany's TUM RoboGym is set to become the world's largest humanoid robot training center. And Lawrence Berkeley Lab turned thermodynamic noise — normally the enemy of computing — into a functional AI asset that could slash inference energy costs dramatically.

🔊 Industry Insights & Updates

China's XGSynBot unveiled the Z1, a wheeled humanoid that switches tools in under six seconds and is built for real industrial grime from day one. AIKO Energy launched its Gen 3 ABC solar panels in Australia, peaking at 545W with less than 0.35% annual degradation after the first year. Lucid Motors revealed Lunar, a steering-wheel-free robotaxi concept built on its upcoming Midsize platform, with Uber already in advanced deployment talks. And the US Department of Energy approved the safety documentation for the MARVEL microreactor, clearing the path for the first nuclear physics trials at Idaho National Laboratory.

🧬 BioTech

A RIKEN team in Japan identified the missing third cell type needed to grow a complete, cycling hair follicle in the lab — follicles that produced hair, cycled repeatedly, and even connected to host nerves after transplantation. At Baylor University, researchers bonded the toxin saporin onto Listeria bacteria, using the pathogen's natural ability to invade cells as a drug delivery vehicle to kill colorectal cancer cells from the inside. And in a historic telesurgery milestone, surgeon Prokar Dasgupta successfully removed a patient's prostate from 2,400 km away, guided by the Toumai Robotic System with just 48 milliseconds of lag.

💡 Products/Tools of the Week

Triall is an AI verification platform that runs three independent large language models in parallel, stress-testing each other's outputs through blind peer review and adversarial refinement — delivering a single evidence-tagged verdict with confidence scores for legal, technical, and research professionals who can't afford a hallucination. Hebbrix gives AI agents genuine long-term memory through a 3-tier architecture and 5-layer hybrid search, retrieving context in under 50ms and plugging into any LLM stack via OpenAI-compatible APIs. Image Describer turns any uploaded image into SEO captions, WCAG-compliant alt text, and Midjourney-ready prompts in seconds, with batch processing built in. And Dashtera is a GPU-accelerated dashboarding platform that visualizes billions of data points in real time with millisecond-level updates — no code required.

🎥 Video Section

Figure's Helix 02 tidies a living room with an ease that genuinely unsettles. DEEP Robotics built a robot horse that moves with a fluidity that has no business existing yet. And BIGAI's OmniXtreme robot delivers agility so smooth it looks like CGI — except it isn't.

The most exciting thing about this moment isn't any single breakthrough — it's that all of them are happening at once. Brains being decoded, robots learning to feel, surgeons crossing continents, and atoms being split in microreactors the size of a room. The next decade won't just look different — it will be unrecognizable. Stay hungry, stay futurish!

🤯 Mind-Blowing

• Videos were reconstructed from mouse brain signals alone by researchers at University College London, marking a significant step forward in neural decoding. Dr. Joel Bauer, principal investigator at the Sainsbury Wellcome Centre at UCL, led the team in using single-cell recordings from the visual cortex — rather than fMRI machines — to build a dynamic neural encoding model that predicts how individual neurons respond to specific video frames. The model also factored in the mouse's physical state, including movements and pupil dilation, to better reflect how internal conditions shape perception. Reconstructed clips ran to 10 seconds, and their accuracy improved as more individual neurons were tracked.

• Grip sensitivity matching human hands was demonstrated by FORTE, a new robotic hand built at the University of Texas at Austin that successfully handled fragile items like potato chips and raspberries without damage. Siqi Shang, the doctoral candidate who led the research, said the key innovation is the use of 3D-printed fingers modeled on the fin-ray effect from fish fins, with internal air channels that function as pressure sensors and deliver real-time force feedback. Shang's team tested FORTE on 31 different objects and achieved a 91.9 percent grasping success rate, while the system accurately flagged 93 percent of slip events without a single false alarm. Assistant professor Lillian Chin noted that FORTE's sensors respond at speeds comparable to human hand receptors — a benchmark no prior robotic gripper had reached.

• A robot AI that predicts physical motion from video and uses those predictions to guide machines in real time was unveiled by Rhoda AI as the company emerged from stealth. Called FutureVision, the system runs a Direct Video Action model that generates short video forecasts of what will happen next in the physical environment — every few hundred milliseconds — and immediately translates those forecasts into robot movements. Rhoda CEO Jagdeep Singh said the breakthrough is that robots no longer need exhaustive pre-programming for every scenario; instead, FutureVision gives them the ability to anticipate and adapt continuously. In a production trial, a robot guided by FutureVision completed a component-processing task in under two minutes per cycle with no human assistance.

• A massive robotics facility is set to open in Germany, built to become the world's largest training center for humanoid robots. Known as TUM RoboGym, the center is a joint venture between the Technical University of Munich and NEURA Robotics, a firm based in Metzingen, with a combined investment of approximately €198 million — €128 million of which comes from NEURA Robotics alone. Located near Munich Airport at the TUM Convergence Centre, the facility will span around 25,000 square feet and house hundreds of robots learning everyday tasks from human trainers. David Reger, founder and CEO of NEURA Robotics, stated that the competitive advantage in intelligent robotics is no longer hardware, but high-quality training data.

• Thermal noise, the random electron movement that conventional computers spend enormous energy suppressing, has been turned into a computing asset by researchers at Lawrence Berkeley National Laboratory. The team, including Molecular Foundry scientist Whitel and researcher Corneel Casert, created a framework that enables thermodynamic computers to handle nonlinear AI tasks — something previously out of reach for the field. Casert trained the system by running evolutionary simulations across 96 GPUs on the Perlmutter supercomputer, screening more than a trillion noisy trajectories using a genetic algorithm. Once deployed as physical hardware, the approach promises dramatically lower energy use for AI inference compared to conventional processors.

🔊 Industry Insights & Updates

• Switching tools in under six seconds, the Z1 humanoid robot was introduced by China’s XGSynBot as a direct answer to the gap between agile robotics and genuinely tough industrial environments. Revealed at the company’s “More Than One Answer” launch event in both Silicon Valley and Beijing, the Z1 is a wheeled humanoid equipped with the world’s inaugural Modular-End-Effector Quick Change System and in-house XG-High-Performance Joint Modules. XGSynBot’s CEO described the robot as a “blue-collar worker” built for oil-stained factory conditions from the first day of deployment, not a showpiece confined to controlled settings. Alongside the Z1, XGSynBot announced the STARFIRE ecosystem to open its hardware interfaces to third-party developers and progressively release proprietary SDKs to the broader research community.

• Releasing its most advanced panel to date, AIKO Energy on March 11 introduced the Gen 3 ABC 60-Cell module in Australia, a product designed to squeeze more power from limited suburban rooftops. The module's Infinite ABC technology eliminates metal grid lines from the front surface, creating a pure black light-absorbing panel that peaks at 545W — enough to push a typical 660m² commercial factory roof from 100kW to 107kW using the same footprint as before. Degradation is minimal: just 1 percent in the first year and 0.35 percent annually thereafter, ensuring over 90.6 percent power retention after three decades. Bywater of AIKO Energy said the uniform format and installation process across all project types allows installers to scale their businesses more efficiently.

• A two-seat robotaxi with no steering wheel and no pedals was unveiled by Lucid Motors at its investor day event in New York. Named Lunar, the concept is built on Lucid's upcoming Midsize electric vehicle platform — the same architecture that will underpin two consumer SUVs, the Lucid Cosmos and Lucid Earth, with vehicles starting under $50,000. Lucid is in advanced negotiations with Uber to deploy vehicles from this platform at scale, with Uber CEO Dara Khosrowshahi calling Lucid "a vital strategic partner" for rolling out autonomous vehicles globally. Interim Lucid CEO Marc Winterhoff said the company is applying greater scale and cost discipline while preserving its core technology DNA.

• Safety approval for the MARVEL microreactor's foundational documentation was granted by the US Department of Energy, clearing the path for the first nuclear physics trials at Idaho National Laboratory. The compact reactor uses sodium-potassium cooling and is engineered to deliver between 85 and 100 kilowatts of thermal output alongside approximately 20 kilowatts of electrical power, making it one of the smallest grid-capable nuclear units ever designed. INL's Alla Ab-Jade described the approved safety framework as a reusable model for advanced nuclear developers seeking to speed up their own programs. The risk-informed approach used by MARVEL has already begun to influence other DOE-backed projects, including the Molten Chloride Reactor Experiment and VALKR.

Subscribe now

🧬 BioTech

• Scientists in Japan have cracked a long-standing puzzle in hair regeneration by identifying the third essential cell type needed to grow a complete, cycling hair follicle in a lab. The RIKEN team — including Takashi Tsuji, Koh-ei Toyoshima, and Miho Ogawa — showed that dermal mesenchymal cells marked PDGFRα+/Sca1+/CD34high+ act as the structural scaffolding of the growing follicle, differentiating into dermal sheath cells that physically wrap around and pull the follicle downward as it grows. Without this population, engineered follicles formed hair bulbs but could never elongate into full follicles — the equivalent of building a foundation without ever raising the walls. With all three populations in place in a 3D skin model, the follicles produced hair, cycled repeatedly, and even connected to host nerves and muscles after transplantation.

• Colorectal cancer cells were killed more effectively than ever before in mouse models by a bacteria-based drug delivery system developed at Baylor University and published in Cell Chemical Biology. Professor Michael VanNieuwenhze and his team — including doctoral students Wyatt Paulishak and Jianan Lyu and a collaborator from Texas Tech University Health Sciences Center — chemically bonded the toxin saporin onto the surface of Listeria monocytogenes, a bacterium already famous for its ability to slip inside human cells undetected. Once Listeria invades a tumor cell, the saporin is released into the cytoplasm, where it shuts down the cell's protein-making machinery and kills it. VanNieuwenhze described the approach plainly: "hook saporin on the surface of a bug, let the bug get delivered into the cell, and use chemistry inside the cell to release saporin to kill it."

• Remote surgery reached a new milestone when Prokar Dasgupta, head of robotic surgery at The London Clinic, successfully removed a patient's prostate from 2,400 kilometers away. Paul Buxton, a 62-year-old cancer patient at St Bernard's Hospital in Gibraltar, had originally expected to travel to the UK for treatment but instead took part in a telesurgery trial. Dasgupta guided the Toumai Robotic System in real time through a secure high-speed network provided by Presidio, achieving a lag of only 48 milliseconds between his London console and the robot in Gibraltar. A local surgical team was present at Buxton's side throughout the operation.

  ---

💡Products/tools of the week

• Launched to address the growing demand for auditable AI outputs, Triall is an AI-driven verification platform that runs three independent large language models in parallel, subjecting each model's answers to blind peer-review, stress-testing, and adversarial refinement by the others. The system then performs convergence analysis and devil's-advocate checks, topped by web-based claim verification, before delivering a single evidence-tagged verdict complete with confidence and over-compliance risk scores. Professionals in legal, technical, strategic, and research fields turn to Triall when a single model's output is simply not reliable enough, as its multi-model workflow is specifically designed to catch correlated hallucinations and leave a transparent reasoning trail that solo AI systems cannot match.

• Hebbrix is a persistent memory platform that gives AI agents and applications long-term, queryable context through a 3-tier memory architecture spanning short, mid, and long-term storage. The platform combines a 5-layer hybrid search — covering vector similarity, BM25, knowledge-graph traversal, time-decay, and ONNX reranking — with a knowledge graph engineered for temporal reasoning and contradiction detection, all at sub-50ms retrieval speeds. Developers can plug Hebbrix into any LLM stack via OpenAI-compatible APIs to build chatbots, agents, and AI apps that remember, reason, and personalize over time, without constructing custom state-management or memory systems from scratch.

• Released for creators, marketers, and accessibility teams who need image copy without the manual effort, Image Describer is an AI-powered tool that analyzes uploaded images and produces ready-to-use natural-language outputs instantly. Using computer vision and language models, it detects objects, scenes, mood, style, and composition to deliver short or detailed descriptions, WCAG-compliant alt text, SEO-friendly captions, OCR-extracted text, and image-to-prompt formats for Midjourney and Stable Diffusion. Image Describer supports batch processing and interactive image chat, giving teams a fast, scalable way to generate copy, prompts, and accessibility tags without touching a single line of manual description work.

• Built for engineering, finance, IoT, healthcare, and other data-intensive domains, Dashtera is a GPU-accelerated, real-time dashboarding platform that visualizes massive, high-frequency datasets — from millions to billions of points — with smooth 2D and 3D charts and millisecond-level updates. The platform connects to SQL and streaming sources, supports file imports, and lets teams build and share interactive dashboards without writing a single line of code. Dashtera also includes anomaly detection, data-transformation tools, and ML/AI workflow integration to surface automated insights and alerts wherever ultra-low-latency observability is critical.

  ---

• Figure: Helix 02 Living Room Tidy

• DPCcars: DEEP Robotics Built a Robot Horse and It Is Incredible

• BIGAI:🤸‍♂️🤖 INSANE AGILITY! #OmniXtreme Robot Shreds

Welcome to this week's edition of Rushing Robotics! The future is accelerating, and this week’s discoveries blur the line between biological intuition and machine capability. From emotional androids to molecular time capsules, we’re witnessing the convergence of sensing, storage, and cognition at scales that feel almost magical.

🤯 Mind-Blowing

This week's stories close the gap between human and machine faster than most people realize. China built one of the largest 3D facial databases ever assembled so robots can genuinely understand emotions. Electronic skin inspired by the human pupil switches between long-range detection and feather-light touch sensing on demand. Quantum computing is being retooled from the ground up for data problems classical AI can't crack. A Swedish team proved seasonal solar storage works in a real apartment building. And a German physics team built an OLED pixel so small a full HD display fits within a single square millimeter.

🔊 Industry Insights & Updates

The energy and robotics industries are having a serious moment. Aptera rolled the first solar EV off a real production line, backed by over $2 billion in reservations. A German university team created a solar battery that stores sunlight for days and releases hydrogen on command at 80 percent efficiency. Noble Machines emerged from stealth with a record 1,000-robot deployment at a single Texas site, continuously inspecting infrastructure around the clock. And Argonne's Aurora supercomputer is compressing days of fusion plasma simulation into hours, with AI models that can flag a dangerous reactor disruption in milliseconds.

🧬 BioTech

Biology is getting a serious upgrade this week. Researchers built a soft-printed synthetic heart that actually beats, giving cardiac surgeons a realistic environment to rehearse complex procedures before touching a real patient. Rockefeller University completed a sweeping atlas of nearly 7 million cells across 21 organs, revealing that aging begins earlier and runs more synchronized across the body than anyone expected. And in Missouri, researchers cracked DNA storage's biggest limitation — a fully rewritable molecular hard drive that encodes, erases, and re-encodes data, stable at room temperature with no power required.

💡 Products/Tools of the Week

This week's tools are built for people who want to move faster without the friction. Town connects all your communication and productivity apps into one AI assistant that handles correspondence, scheduling, and workflows — no coding required. Pencil merges design and development into a single IDE canvas, letting designers and developers work side by side with AI generating layouts instantly. DittoDub automates video dubbing across languages while cloning the speaker's voice and keeping lip sync intact. And QuiverAI turns text prompts or raster images into polished, editable SVGs — a game changer for anyone tired of manual tracing.

🎥 Video Section

Before you go, this week's videos are worth every second — humanoid robots hitting real factory floors at BMW and Xiaomi, AGIBOT's next-gen industrial machine, and HONOR revealing a robot phone that's apparently shipping, not just a concept.

The pace of change isn't slowing — it's compounding. Every week, yesterday's breakthrough becomes today's baseline, and what felt impossible just months ago is already being deployed at scale. Stay hungry, stay futurish!

🤯 Mind-Blowing

• A massive 3D facial database has been built in China to help humanoid robots and virtual humans better read and display emotions. Researchers led by Professor Song Zhang at the Shenzhen Institutes of Advanced Technology and collaborators at Xidian University created an AI model that learns directly from raw 3D face scans instead of relying on 2D images or pre-made templates, which often misalign with real human anatomy. Their new curvature-fused graph attention network (CF-GAT) analyzes unordered point clouds, preserving subtle curvature details while still understanding the global structure of each face. The team gathered around 200,000 high-fidelity scans, covering multiple expressions, precise landmarks, and even dynamic 4D facial movements, forming one of the largest real 3D human face datasets ever assembled. This work aims to make androids, biometric systems, and virtual avatars far more lifelike and robust in real-world conditions.

• An electronic skin that thinks like an eye has been engineered by researchers at South China University of Technology to give robots both distant awareness and a gentle touch. The team designed a flexible capacitive sensor array with a dynamic shielding layer that mimics how the human pupil expands and contracts, reconfiguring electrode coverage to switch between long-range proximity detection and fine tactile sensing on demand. In proximity mode, the eye-inspired electronic skin detected objects from beyond 90 mm, more than doubling the range of conventional dual-mode sensors, while in touch mode it registered forces as light as a few grams and endured pressures up to 400 kPa. Published in the International Journal of Extreme Manufacturing, this work positions the adaptive electronic skin as a single unified solution for collaborative robots navigating shared spaces and handling fragile objects.

• Data-scarce AI problems became the focus of a new quantum research push when Xanadu and Lockheed Martin announced their collaboration on quantum machine learning generative models. Xanadu, the quantum computing company behind the PennyLane software framework and photonic hardware, will work with Lockheed Martin to investigate Fourier-based and quantum-native learning methods that classical generative models like large language models and image generators cannot replicate with similar efficiency. Lockheed Martin's quantum technologies lead Dani Couger highlighted that the research could transform computation and sensing to support national security and advanced system development. Xanadu CEO Christian Weedbrook said the goal is to revisit quantum primitives from the ground up, with the aim of discovering fundamentally new methods for data representation and processing that could influence hardware and algorithm development for years to come.

• Excess summer sunlight has been turned into winter heat in a real apartment building, showing seasonal solar storage can work at scale. Researchers from Chalmers University of Technology in Sweden developed a reversible phase change material called a molecular solar thermal energy storage system (MOST) that stores solar energy in liquid form for up to 18 years, then releases it later as heat on demand. In their latest test, they integrated MOST into a centralized heating system in a multi-family building and successfully delivered stored summer energy to residents during the colder months, reducing reliance on conventional heating and cutting emissions. The team now aims to refine efficiency, integrate with district heating, and explore pairing the technology with photovoltaics for combined electricity and heat solutions.

• The smallest OLED pixel ever built has been created by physicists at Julius-Maximilians-Universität Würzburg in Germany, measuring just 300 by 300 nanometers while matching the brightness of conventional pixels 270 times its size. Professors Jens Pflaum and Bert Hecht led the team in solving a long-standing problem: at nanoscale, electrical current concentrates at the corners of the antenna like a lightning rod effect, causing gold atoms to grow filament-like threads into the active material until a short circuit destroys the device. Their solution was a precisely engineered insulating layer with a 200-nanometer circular opening at the center, blocking edge currents and keeping the nanopixel stable for over two weeks under normal conditions. The result means a full HD 1920 x 1080 display could fit within one square millimeter, potentially allowing projectors to be built directly into eyeglass arms, with efficiency improvement and full RGB color range as the team's next targets.

🔊 Industry Insights & Updates

• A solar electric vehicle has rolled off an assembly line for the first time at Aptera's California production facility, signaling the startup's transition from development into manufacturing. Aptera, the San Diego-based solar mobility company, built a 14-station validation line staffed by dedicated vehicle line technicians to ensure repeatable builds and quality assurance ahead of certification. The vehicles produced will be used for thermal, brake, and destructive evaluations required for self-certification and EPA endorsement. Aptera has also expanded its workforce, making manufacturing its largest team as the company moves from engineering development into testing and production execution. Customer deliveries are targeted for later in 2026, backed by a reservation backlog exceeding $2 billion.

• A solar-powered battery that stores sunlight for days and releases green hydrogen on demand has been developed by researchers at Ulm and Jena universities in Germany. The team built the device around a water-soluble copolymer with high redox activity, which absorbs solar energy and holds a charge for several days before releasing hydrogen when an acid and a catalyst are introduced. The system achieves 80 percent efficiency during charging and 72 percent efficiency during hydrogen release, and the copolymer can be recharged simply by returning it to sunlight, enabling repeated storage and release cycles. A pH switch controls the entire process and also acts as a visual indicator, with the material turning violet when charged and yellow when discharged. Published in Nature Communications, the research merges polymer chemistry and photocatalysis in a way its authors describe as a step toward affordable, scalable solar storage.

• Autonomous operations scaled rapidly when Noble Machines revealed it had installed a fleet of 1,000 S1 and S2 industrial field robots at a single industrial customer site in Texas, touting it as the largest one-site rollout in its history. The company, focused on rugged outdoor robotics, designed the deployment so the robots continuously roam and inspect infrastructure, using advanced sensing and AI-driven analytics to catch anomalies that human teams might miss on periodic rounds. Information captured by the robots is funneled into Noble Machines’ software platform, where it is analyzed, prioritized, and integrated into maintenance and safety workflows. With this record deployment, Noble Machines is positioning robotics fleets as a cornerstone of modern industrial asset management, helping clients tackle labor shortages, enhance safety, and maintain constant situational awareness in the field.

• A quintillion calculations per second are now being aimed at one of energy's hardest problems. The Aurora supercomputer at Argonne National Laboratory is simulating fusion plasma behavior inside tokamaks and training AI models to predict dangerous reactor disruptions before they occur. Using Aurora's 20.4 petabytes of memory, researchers are running simulations of billions of plasma particles under the magnetic conditions expected inside ITER, the international fusion reactor under construction in France. What once took days of computing can now be compressed into hours, and the AI disruption models being trained on Aurora can generate a risk score in milliseconds — fast enough for operators to act before a plasma collapse causes serious damage.

Subscribe now

🧬 BioTech

• A beating synthetic heart has been created by researchers to give cardiac surgeons a realistic environment to practice before operating on real patients. The team used soft material 3D printing to reproduce the anatomical structure of the heart’s left side, including the ventricle and mitral valve, and embedded McKibben actuators in the muscle walls to mimic the contractions of a living heart. Sutures connecting the ventricle to the mitral valve replicate the chordae tendineae, the natural tendon-like cords that keep the valve functioning properly, while custom flexible pressure sensors provide real-time feedback on blood flow dynamics. The model is designed specifically to simulate edge-to-edge repair, a surgical technique used to fix leaking atrioventricular valves, allowing physicians to rehearse complex procedures before entering the operating room.

• A sweeping cellular census of the aging body has been completed by Rockefeller University scientists who mapped nearly 7 million cells across 21 organs to reveal that aging is far more synchronized and earlier-starting than previously understood. Professor Junyue Cao and graduate student Ziyu Lu used refined single-cell ATAC-seq technology to track cell populations in mice from young adulthood through old age, finding that some cell types had already begun declining by five months, which Cao described as a continuation of developmental processes rather than a late-life event. The atlas identified around 300,000 genomic regions showing aging-related changes, with roughly 1,000 of those alterations appearing consistently across many different cell types, suggesting shared biological programs drive aging body-wide. Around 40 percent of aging-associated changes varied by sex, with females showing broader immune activation, which Cao speculated could help explain the higher prevalence of autoimmune diseases in women.

• Achieved for the first time at the molecular scale, a fully rewritable DNA hard drive has been created by researchers at the University of Missouri, redefining what is possible in long-term data storage. For years, DNA has been recognized as an incredibly dense and durable medium, but existing approaches treated it like a write-once archive where information, once encoded, could not be changed. In this new system, the team led by Professor Qun “Andrew” Gu of chemical and biomedical engineering developed a method to encode, erase, and rewrite digital data directly into DNA strands multiple times, much like how files are managed on a traditional hard drive. Paired with a nanopore-based reader that converts molecular signals back into digital bits, the platform can store massive volumes of information at room temperature without constant power. By combining DNA’s stability with true rewritability, the work points toward compact, long-lived “molecular drives” that could safeguard critical personal, scientific, and institutional data for generations.

  ---

💡Products/tools of the week

• Created to streamline digital work, Town acts as an integrated AI assistant connecting with communication and productivity apps like Slack, email, calendars, and file storage. It manages correspondence, schedules, document generation, and workflow automation while maintaining contextual understanding across tasks. Its design balances autonomy and transparency through audit logs, defined permissions, and approval gates, empowering individuals and organizations to automate without coding while maintaining human oversight.

• Pencil integrates a design canvas inside the IDE so developers design and code within one interface. Every project lives in an open, version-controlled repository, allowing standard branching and merging. AI agents help generate layouts and components instantly, while designers and developers can work concurrently. With support for Figma files, editable CSS, built-in kits, and ready React exports, Pencil fully connects to APIs and data sources, giving users direct read/write access to all design files.

• Designed for global content creators, DittoDub automates video dubbing and localization by cloning the speaker's voice and adapting it across multiple languages without losing authenticity. Developed by its founding team, the platform uses AI-driven speech recognition, generative synthesis, and timing-aware audio placement to keep lip sync intact. Subtitles, metadata, and thumbnails are translated alongside the audio. A built-in human review layer ensures quality before delivery, making it a cost-effective alternative to manual dubbing studios.

• Released to close the gap between written ideas and finished vector assets, QuiverAI translates text prompts and raster images into polished, editable SVGs using models built from the ground up for vector design. The platform supports raster-to-vector conversion, streaming progressive rendering, and full animation capabilities. QuiverAI's team designed it so both designers and developers can prototype faster, eliminate tedious manual tracing, and integrate intelligent vector generation into existing design pipelines through a straightforward API.

  ---

• BMW Group: Can a humanoid robot be integrated under real industrial conditions? Our BMW iFACTORY approach 🦾

• XRoboHub: 🔥🤖 Xiaomi Humanoid Robot HITS the factory! 100% Real nut assembly!

• AGIBOT: AGIBOT G2: The Next Generation of Industrial-Grade Interactive Embodied Operation Robots

• HONOR: HONOR Robot Phone. Prototype? No. Reality.

Welcome to this week's edition of Rushing Robotics - your weekly passport to the breakthroughs, innovations, and ideas reshaping our world. This week’s issue is packed with stories that blur the line between science fiction and reality: bacteria engineered to eat cancer from the inside out, robot swarms smaller than a grain of salt moving objects tens of thousands of times their own weight, and a fully functional electric motor printed in three hours for less than a dollar.

🤯 Mind-Blowing

This week’s mind-blowing picks are a reminder that the most transformative ideas often sound impossible right up until they aren’t. From living bacteria programmed to hunt tumors, to quantum-inspired robots navigating crowded spaces in real time, science is operating at a pace that makes last year’s breakthroughs feel almost quaint.

🔊 Industry Insights & Updates

The robotics and energy industries are moving fast — and this week's updates show just how quickly lab ideas are turning into deployments at scale. Toyota is putting humanoid robots to work on factory floors, a tofu-brine battery is logging 120,000 charge cycles, and Honor is about to unveil a humanoid robot at MWC. The commercial era of robotics isn't coming. It's here.

🧬 BioTech

Biology continues to rewrite what we thought was fixed. This week: a baby born through a deceased-donor womb transplant, a bacterial gene hiding inside plants that could unlock entirely new drug pipelines, and a single-dose hepatitis B therapy showing immune responses that last up to a year. Each story is a quiet revolution.

💡 Products/Tools of the Week

The best tools don't just solve problems — they eliminate entire categories of friction. This week's picks do exactly that: Fusedash turns raw data into interactive dashboards and charts in minutes using natural-language chat, Owlytics.ai compresses weeks of market research into a single structured report, AIArtist converts a text prompt into a polished motion-graphics video in under three minutes, and FurniMesh reduces a full 3D modeling workflow to a single photo upload. Whether you're a founder validating an idea, a creator building content, or a designer shipping faster — there's something here for you.

🎥 Video Section

Sometimes you just need to see it to believe it. This week's videos bring the future to life — from Unitree's companion robot to a strawberry-picking machine improving its own harvest rate.

We are living through a moment of compounding breakthroughs — where advances in biology, robotics, materials science, and AI are no longer happening in isolation but feeding into one another at an accelerating pace. The stories in this issue alone hint at a world where cancer is fought from within, where robots shoulder our most demanding labor, and where energy storage finally matches the ambition of renewable generation. Stay hungry, stay futurish!

🤯 Mind-Blowing

• A method of destroying cancer tumors from the inside out using living bacteria has been engineered by a team at the University of Waterloo. Clostridium sporogenes, a bacterium naturally suited to oxygen-free environments, is directed by researchers Dr. Aucoin and Dr. Ingalls to invade the dead, airless cores of solid tumors, where it multiplies and consumes the cancer. The challenge of the bacteria dying off before reaching the tumor’s oxygen-exposed edges was solved by inserting a gene that boosts oxygen tolerance, controlled by a quorum sensing switch that only activates when a sufficient bacterial population has built up. Dr. Ingalls described the control system as a DNA-based electrical circuit in which each genetic piece performs a specific function. The collaboration also features Dr. Sara Sadr, a former Waterloo doctoral student whose work was central to advancing the research, and is being pursued further through CREM Co Labs in Toronto.

• A first-of-its-kind robot cluster has been deployed across the Hefei metro system in China, combining humanoid robots, robotic dogs, autonomous inspection units, and drones into a unified intelligent dispatching platform. Rolled out during one of the year's peak travel seasons in Hefei, the capital of Anhui Province, the system operates across three primary domains: passenger services inside stations, vehicle inspection, and tunnel inspection. Humanoid robots assist travelers with navigation and transfer information, robotic dogs patrol platforms for safety, and autonomous robots crawl through 1.5-meter-deep maintenance trenches to scan wheels, bolts, and mechanical components using high-definition cameras and ultrasonic sensors. Dai Rong of Hefei Rail Transit said the system is designed to assist human staff, reduce work intensity, and improve operational safety rather than replace workers. A supervisor at Hefei's Science and Technology Center added that future development aims to give the robots a shared "central brain" using AI, enabling them to identify and respond to situations more accurately.

• A fully working electric motor was 3D printed in about three hours by researchers at MIT, using a single multimaterial platform that could one day allow factories to produce replacement motors on-site without waiting on global supply chains. The team, led by Fernando Velásquez-García, principal research scientist at MIT's Microsystems Technology Laboratories and lead author of the study, enhanced an existing extrusion-based 3D printer with four specialized extruders — each handling a different material type, from melted filaments to conductive inks. The motor required just one post-printing step: magnetizing its hard magnetic elements, with raw material costs estimated at around 50 cents. Velásquez-García described the achievement as a proof of concept with far broader ambitions, envisioning a future where complex electronic and electromechanical systems can be made locally in a single build.

• Swarms of robots the size of a grain of salt have moved objects 45,000 times their own weight without touching them, in a study conducted by scientists from the Max Planck Institute for Intelligent Systems, the University of Michigan, and Cornell University. Each microrobot measures just 300 micrometers and, when exposed to an external magnetic field, spins to generate tiny whirlpools in the surrounding fluid; when hundreds operate together, those individual currents combine into a powerful force called fluidic torque. Using this contactless approach, the swarms rotated gears, transported structures, and assembled configurations far larger than any single robot could physically handle. Researcher Steven Ceron of the University of Michigan noted that while hydrodynamic drag has historically hindered microbot collectives, this work deliberately harnesses fluid interactions to manipulate objects from a distance. Published in Science Advances, the research points toward future applications in micro-manufacturing and non-invasive medical procedures, including drug delivery and implant assembly inside the human body.

• Quantum-inspired computing has been embedded directly into a mobile robot for the first time, with Japanese companies Toshiba and MIRISE Technologies achieving real-time autonomous control without the need for a remote server. Toshiba's Simulated Bifurcation Machine (SBM) — an optimization system that runs on standard hardware like FPGAs rather than specialized quantum equipment — was mounted on a MIRISE-engineered autonomous platform and tasked with tracking multiple moving objects simultaneously in complex, crowded environments. The system achieved 23 frames per second, more than doubling the 10 FPS threshold typically required for autonomous driving, while delivering a 4 percent improvement in overall tracking accuracy and a 23 percent improvement on metrics measuring performance when objects are obscured. In physical tests, the robot successfully navigated around multiple moving obstacles by using SBM tracking data to assess object positions, predict movement, and plan paths more efficiently. Toshiba and MIRISE plan to extend the technology to autonomous vehicles, multi-robot coordination, and real-time task management across industries.

🔊 Industry Insights & Updates

• Outlasting lithium-ion batteries by a factor of ten or more, a tofu-brine battery developed by researchers from the University of Hong Kong and the University of Science and Technology has logged over 120,000 charge cycles — a figure that could translate to decades of useful life in grid-scale storage applications. The innovation lies in its organic electrodes and neutral electrolyte derived from the saline byproduct of tofu production, which makes the battery non-flammable, non-toxic, and safer to dispose of than conventional lithium-based cells. Lithium-ion batteries, by comparison, are known for thermal runaway risks when damaged and typically degrade within 1,000 to 3,000 cycles for most consumer applications. The team published their findings in Nature Communications, stating the system shows "remarkable long-term cycling stability and environmental sustainability under neutral conditions." Analysts note that transitioning from lab results to commercial-scale production remains the critical challenge before the technology can be deployed in solar farms, wind energy balancing, or rural electrification.

• Robotic hands that can replicate human movement and grip have been built by Chinese startup LinkerBot, founded by Alex Yong, in a development that brings humanoid robots closer to practical everyday use. The company recently secured nearly $217 million in a Series B funding round, which it plans to use to double its research team by the end of 2026. LinkerBot's entry-level O6 model features 11 degrees of freedom and a grip force of 50 kg while weighing just 370 grams, while the flagship L30 model achieves precision within ±0.2 mm for delicate assembly work. To support skill development across robots, LinkerBot also launched LinkerSkillNet, a repository of around 500 reusable skills — from industrial tasks to medical procedures — with CEO Zhou expecting that number to double every six months. Partners include Stanford University, and the company has already delivered its 10,000th dexterous hand.

• Remote control of robots with near-human precision over a commercial 5G network has been demonstrated for the first time by DOCOMO and Haption, who used a 5G feature called Configured Grant to eliminate the scheduling delays that previously made wireless teleoperation unreliable. The test, announced effective February 25, 2026, paired DOCOMO's low-latency network slicing with Haption's Virtuose system — a platform that synchronizes position, velocity, and force in real time between an operator-side robot and a remote one. Results showed a 40 percent increase in force-feedback reproduction and a 59 percent improvement in motion smoothness, as measured by the Dimensionless Jerk Cost metric. Unlike the conventional Dynamic Grant method, where a device must request transmission resources before sending data — creating variable delays — Configured Grant pre-assigns uplink resources so the robot can transmit instantly. DOCOMO and Haption plan to continue refining the system for real-world deployment of teleoperated robots.

• Seven humanoid robots will be put to work on the RAV4 assembly line at Toyota Motor Manufacturing Canada's plant, following a deal between TMMC and Agility Robotics that was confirmed after a February 19 trial. The Digit robots, built by the Oregon firm, will operate under a Robots-as-a-Service model and focus on unloading auto parts from automated warehouse vehicles — repetitive, physically demanding work that currently falls on human workers. TMMC President Tim Hollander stated the company assessed several robotic options before choosing Digit, with the goal of both improving worker experience and increasing manufacturing efficiency. Agility Robotics CTO Pras Velagapudi has previously noted that deployment costs can far exceed the price of the robots themselves, and that AI-driven tools like the Agility Arc platform are key to bringing those costs down.

• A humanoid robot is set to be unveiled by Honor at Mobile World Congress 2026 in Barcelona, opening March 1, marking the first time a Chinese smartphone maker has entered the consumer-grade humanoid robot segment. The robot, developed by Honor, is designed for retail and home service scenarios, with capabilities including product delivery and household companionship. At the same event, Honor will also introduce its ROBOT PHONE, a robotic smartphone featuring a hidden mechanical arm that automatically frames shots and tracks subjects, combining AI-powered imaging with physical movement.

Subscribe now

🧬 BioTech

• A baby boy has been born in the UK to a mother who received a womb transplanted from a deceased donor, marking the country's first successful birth of this kind. Grace Bell, who is in her 30s and was born without a functional uterus due to Mayer-Rokitansky-Küster-Hauser (MRKH) syndrome, underwent a 10-hour transplant surgery at The Churchill Hospital in Oxford before conceiving through IVF. Her son Hugo was delivered at Queen Charlotte's and Chelsea Hospital in London, representing a landmark moment in a UK clinical trial exploring whether womb transplants can become a standard authorized procedure. The transplant was made possible by an anonymous deceased donor, whose family noted she also donated five other organs that helped four additional recipients.

• A bacterial-like gene at the heart of a powerful plant chemical has been discovered by researchers at the University of York, in a finding that could reshape how new drugs are found and made. Scientists studying Flueggea suffruticosa, a plant that produces the alkaloid securinine, found that the gene driving its production looks far more like a bacterial gene than a plant one. Dr. Benjamin Lichman and his team — including Catharine X. Wood, Zhouqian Jiang, and colleagues — believe plants are essentially borrowing and repurposing microbial tools to build their own defensive chemistry. Once the team recognized this new pathway, they identified similar bacterial-like genes tucked into the DNA of many other plant species, suggesting the strategy is widespread.

• Durable HBV-specific immune responses lasting up to one year were reported by Virion Therapeutics following a single dose of VRON-0200 in its ongoing Phase 1b study, with findings shared at the 33rd CROI conference in Denver. Virion Therapeutics described the therapy as the first new HBV immune modulator since pegylated interferon to demonstrate such lasting clinical activity. Dr. Sue Currie, Virion's COO, presented the data and outlined the company's "Spark and Fan" strategy — in which VRON-0200 primes the immune response and antiviral agents amplify it — as a potential backbone for a wide range of future functional cure strategies.

  ---

💡Products/tools of the week

• Launched as an AI-powered solution, Fusedash transforms how teams interact with data by automatically connecting CSVs, APIs, and public datasets to generate charts, maps, dashboards, and real-time KPI views. The platform's AI chart generator recommends chart types and styling so analysts don't have to start from scratch. A built-in natural-language chat lets anyone ask questions and receive charts and explanations instantly, making code-free reporting accessible to every team member.

• Built for founders and growth teams who can't afford to wait weeks for market intelligence, Owlytics.ai is an AI-powered market research platform that runs a large advanced model across 38 specialized research modules — covering TAM, competitive analysis, customer personas, SWOT, pricing, GTM, financial models, and more — to generate professional, data-enriched reports in minutes. Owlytics.ai delivers structured outputs including metric cards, tables, and strategic recommendations that teams can use directly for investor materials and business validation. Data stays protected under SSL encryption with full GDPR and KVKK compliance, and users pay only for what they need through a flexible per-report credit system.

• Designed for speed and scale, AIArtist is an AI-powered motion-graphics generator that transforms text prompts into polished kinetic-typography videos for Reels, Shorts, and ads through a fully automated multi-stage AI pipeline — no design skills or timeline editing required. AIArtist's pipeline runs LLMs to interpret the nuance of tone, pacing, and emphasis in each prompt, passes those signals to generative models that construct contextual visuals and assets, and hands everything to a video engine that renders smooth, professional animations. Exports arrive as 720 or 1080 MP4 files in 2 to 3 minutes, giving creators and marketers a fast, repeatable path to professional-looking social and ad video at scale.

• Reducing a multi-hour modeling task to a single photo upload, FurniMesh is an AI-powered conversion tool that produces professional, production-ready 3D furniture models with automatic color-coded parts separation, making each component — legs, seat, backrest, drawers — editable and texture-ready out of the box. FurniMesh's furniture-specialized machine learning ensures the resulting geometry, topology, textures, and proportions are accurate enough for real production pipelines, with exports available in GLB, OBJ, SKP, and BLEND formats. The cloud-based platform leverages fast GPUs for quick turnaround and adds AR-enabled web viewers and easy website embedding, targeting manufacturers, retailers, interior designers, and 3D artists who need to speed up their visualization workflows.

  ---

• Unitree Robotics: Unitree Introducing | Unitree As2 Your Companion

• German Research Center for Artificial Intelligence: RoLand: Strawberry picking robot SHIVAA with improved picking rate

• Unitree Robotics: Unitree Kung Fu Bot Pray for Blessings at the Temple of Heaven: Wish Everyone All the Best

Welcome to this week's edition of Rushing Robotics! This week, the boundary between science fiction and science fact got a little thinner. From brain implants inspired by paper art to buildings that literally breathe, the pace of discovery isn't slowing — it's compounding.

🤯 Mind-Blowing

This week's Mind-Blowing section is living up to its name in every sense. A flexible brain implant floats on your neurons without tearing them. A decades-old theory about how your eyes develop was quietly dismantled in a lab. Bacteria are being turned into a building material that gets stronger with age. A full 3D object was printed faster than a camera shutter blinks. And an AI found 500 security vulnerabilities that no human had ever noticed — on its own. Each of these stories, on its own, would have been the headline of the decade a generation ago. This week, they all arrived together.

🔊 Industry Insights & Updates

The infrastructure of tomorrow is being laid down fast. Google just pushed reasoning benchmarks into new territory with Gemini 3.1 Pro. Microsoft is storing data inside glass that could outlast civilization as we know it. A 21-year-old founder just gave the robotics world its open-source moment. And NVIDIA is partnering with the Department of Energy to build nuclear reactors at AI speed. The theme connecting all of it: the tools we use to build the future are themselves being rebuilt from the ground up.

🧬 BioTech

Biology continues to be the most audacious engineering discipline on the planet. Dancing molecules are healing lab-grown spinal cords. A single 10-minute DMT infusion is holding severe depression at bay for months. And scientists in Canada are generating electricity from urine — and making it sound like the most logical thing in the world. The lab bench has never felt closer to the clinic, and this week's breakthroughs make that gap look narrower than ever.

💡 Products/Tools of the Week

This week's tools are all about collapsing the gap between idea and output. Whether you're an architect turning a 3D model into a photorealistic render in seconds, a content team wiring together 150+ AI models without writing a single line of code, or an enterprise deploying autonomous AI agents across your entire stack — the common thread is the same: less friction, more creation. These platforms aren't just saving time; they're quietly redefining what a small team can accomplish.

🎥 Video Section

This week's video picks put the humanoid robotics race front and center — from HMND's industrial ambitions to Unitree's New Year celebrations and Boston Dynamics' latest deployment. Watch them back to back and you'll notice something striking: these machines are starting to move less like robots and more like colleagues.

The future isn't some distant destination — it's being assembled in real time, piece by piece, discovery by discovery. Every week that passes, the world we're building looks less like a prediction and more like a promise. Stay hungry, stay futurish!

🤯 Mind-Blowing

• For the first time, a brain implant designed to float on neural tissue was built by Chinese scientists using kirigami — the same art form behind elaborate paper cutting — and it recorded 700 neurons firing simultaneously when tested in monkeys. Developed at the Chinese Academy of Sciences, the device wraps its electrodes into coil-like spiral shapes that can stretch, compress, and flex with the brain's heartbeat-driven motion rather than tearing at surrounding tissue. A hydrogel layer coats the implant during insertion to further reduce friction and cushion the device against the brain. Fang Ying, senior researcher at the Chinese Institute for Brain Research in Beijing, said the team discovered the electrode retraction problem about four years ago and began searching for a structural solution — one that geometry, not new materials, could provide. The implications stretch from paralysis treatment to speech restoration and cognitive enhancement.

• A decades-old theory about how human eyes develop sharp central vision was overturned by researchers at Johns Hopkins University. Scientists used lab-grown retinal organoids to study the foveola — a tiny pit at the center of the retina responsible for nearly 50% of all visual perception — and discovered that the blue cone cells previously thought to simply migrate away from the region actually transform into red and green cones through a two-step chemical process driven by retinoic acid and thyroid hormones during fetal development. The finding challenges a model that had stood unchallenged in the field for 30 years, and opens the door to lab-grown photoreceptors that could one day restore sight in patients with macular degeneration.

• A building material that breathes, heals itself, and eats carbon dioxide was created by scientists, with results published in Nature Communications showing it stayed alive and active for over 400 days. Researchers engineered a 3D-printable hydrogel packed with living cyanobacteria — the same ancient organisms that first oxygenated Earth's atmosphere — and watched it progressively mineralize from the inside out, growing stronger the longer it lived. The bacteria fix CO₂ through photosynthesis while simultaneously triggering calcium carbonate deposits that reinforce the material's own structure, making it the rare building product that improves with age. With no energy input, no toxic byproducts, and nothing but light and nutrients to keep it running, the team behind the study believes it could fundamentally change how the construction industry thinks about carbon.

• A complete 3D object was printed in just 0.6 seconds by researchers at Tsinghua University in China, shattering previous speed records while maintaining a resolution finer than one-fifth the diameter of a human hair. The team abandoned traditional layer-by-layer printing entirely, instead developing a technique called DISH — Dynamically Integrated Synthesized Holographic light fields — which uses holographic light projected from multiple angles to solidify an entire 3D structure inside a resin chamber almost instantaneously. Their system reached a printing speed of 333 millimeters per second while holding a feature resolution of just 12 micrometers. The approach works whether the resin is stationary or flowing through a channel, opening possibilities for continuous mass production of complex components.

• Over 500 previously unknown high-severity security vulnerabilities were discovered in open-source software by Anthropic's Claude Opus 4.6 with minimal human direction, in findings reported exclusively by Axios. Anthropic's frontier red team gave the model access to Python and standard vulnerability tools — debuggers and fuzzers — but provided no specific instructions or expert knowledge, letting Claude work autonomously. The model identified zero-day flaws ranging from system-crashing bugs to memory corruption vulnerabilities in widely used libraries including GhostScript, OpenSC, and CGIF, all of which were confirmed by Anthropic team members or external security experts. Logan Graham, leader of Anthropic's frontier red team, told Axios the models are already exceptionally proficient at this and will only improve — and that he would not be surprised if AI-driven bug hunting becomes the primary method for securing open-source software in the future.

🔊 Industry Insights & Updates

• A major reasoning upgrade to its flagship AI model was released by Google with Gemini 3.1 Pro, arriving just months after Gemini 3 launched in November and bringing record-breaking scores on some of the hardest benchmarks in the field. On Humanity's Last Exam — a test of advanced domain expertise considered one of the toughest AI evaluations in existence — Gemini 3.1 Pro scored 44.4 percent, beating Gemini 3 Pro's 37.5 percent and surpassing OpenAI's GPT 5.2, which scored 34.5 percent. On the ARC-AGI-2 benchmark, which probes genuinely novel reasoning challenges, the new model scored 771 percent higher than its predecessor. Google has kept API pricing flat — $2 per million input tokens — and is rolling the model out across AI Studio, Vertex AI, the Gemini app, and NotebookLM.

• Conventional storage is losing a war against time — and the answer Microsoft found was to stop using materials that time can destroy. Project Silica, developed at Microsoft Research Cambridge encodes data permanently inside sheets of borosilicate glass using a femtosecond laser that fires four simultaneous beams, burning microscopic voxels through multiple layers of the material at 65 million bits per second — fitting 4.84 terabytes, the equivalent of roughly two million printed books, into a plate the size of a coaster just 2 millimeters thick. Because glass is chemically inert, it withstands temperatures up to 290°C, complete water submersion, and electromagnetic pulses that would wipe every hard drive and magnetic tape on the premises, requiring no power or maintenance for up to 10,000 years. Melissa Terras, a digital heritage specialist at the University of Edinburgh, welcomed the medium but warned that for data to remain truly accessible millennia from now, the instructions and equipment needed to read the glass must be preserved alongside it — a reminder that the technology is only as durable as the institutions that maintain it.

• Humanoid robotics just got its Linux moment. RoboParty, a Beijing-based startup founded in April 2025 by 21-year-old Huang Yi, released its Roboto Origin robot as fully open-source in January 2026 — making it the first complete bipedal humanoid in the world to share its entire technology stack publicly, from structural hardware and supplier lists to motor control software and Sim2Real gap solutions. Built in just 120 days from April to August 2025, the 1.25-meter robot walks at 3 meters per second using a proprietary anthropomorphic gait algorithm and was designed specifically to stress-test full-stack humanoid development with a small team. With $10 million in backing from Xiaomi-affiliated MPCi and a GitHub community already crossing 1,000 stars, founder Huang Yi said his ultimate goal is a world where the best robotic solutions emerge from open collaboration.

• A partnership between the Idaho National Laboratory and NVIDIA was launched to use artificial intelligence to cut nuclear reactor build times in half and slash operational costs by over 50 percent. Named Prometheus, the initiative sits inside the Department of Energy's broader Genesis program and deploys generative AI models, digital twins, and agent-based workflows to automate the engineering tasks that currently stretch reactor development across decades. NVIDIA will provide GPU-accelerated infrastructure while INL contributes historical nuclear data, laboratory findings, and experimental reactor operations to train the digital twin models. INL director John Wagner said the collaboration could fundamentally transform how quickly advanced nuclear energy comes online — an urgent need as AI data centers drive electricity demand to record levels.

Subscribe now

🧬 BioTech

• A lab-grown human spinal cord was healed after injury by researchers at Northwestern University using a therapy called “dancing molecules”. Scientists led by Samuel I. Stupp — the inventor of the therapy and a pioneer in regenerative nanomedicine — grew the most advanced spinal cord organoids ever created from stem cells, incorporated the central nervous system’s own immune cells for the first time, then simulated lacerations and contusions before applying the treatment. The dancing molecules, injected as a liquid that gels into a nanofiber scaffold on contact, rapidly move and vibrate to connect with the brain’s constantly shifting cellular receptors — and the difference was immediate: treated organoids sprouted long neurite extensions while glial scar tissue significantly shrank. The therapy already reversed paralysis in mice in 2021 and recently earned an FDA Orphan Drug Designation, bringing it one critical step closer to human trials.

• A single session with one of the most powerful hallucinogenic substances on Earth wiped out severe depression for up to six months in a clinical trial. Scientists at Imperial College London administered DMT — dimethyltryptamine, the psychedelic compound famous for inducing visions of other dimensions and encounters with alien entities — intravenously to 17 adults with moderate to severe major depressive disorder, while the other 17 received a placebo. Those who received the drug showed dramatically greater improvement than the placebo group, with antidepressant effects lasting three to six months from a single 10-minute infusion. Principal investigator Dr. Erritzoe called the result exciting precisely because of its simplicity — one session, one drug, one course of psychological support, and months of relief.

• Electricity was generated from human urine by researchers at McGill University in Canada, with findings showing that higher urine concentrations produced greater power output from microbial fuel cells. Professor Vijaya Raghavan and his bioresource engineering team built four dual-chamber microbial fuel cells and tested mixtures of wastewater and urine at concentrations of 20%, 50%, and 70% over two weeks, tracking energy output, pollutant removal, and shifts in the microbial communities inside each cell. The system works by allowing bacteria to oxidize organic compounds in the urine at the anode, releasing electrons that travel through a circuit to generate usable electricity. Raghavan suggested the technology could one day power rural communities or disaster-affected regions while simultaneously treating waste — no grid, no infrastructure, no fuel required.

  ---

💡Products/tools of the week

• Speed and consistency define what Eler built for architects and designers. The AI rendering platform takes SketchUp or GLB models and produces photorealistic images in seconds by reading geometry directly, skipping prompt engineering altogether. One batch render locks in materials, lighting, and detail across every angle — cutting both the cost and the wait of traditional visualization workflows.

• CNAPS Studio, launched in early 2026 and backed by NAVER D2SF and Bluepoint Partners, gives nontechnical users a drag-and-drop canvas to connect 150+ pre-trained text, image, video, and data AI models into production-ready workflows. CEO Innfarn Yoo and the CNAPS team built intelligence mapping at its core — the system automatically determines which models to connect and how, balancing output quality against cost. For e-commerce and content teams, that means rapid prototyping without trial-and-error integration cycles or dedicated engineering resources.

• Agentic AI just got an operating system. SimplAI, built by Datafuse Technology, gives enterprises a unified platform to build and deploy autonomous AI agents using drag-and-drop builders, RAG-backed knowledge bases, and multimodal voice interfaces — all without writing code. The platform orchestrates models across OpenAI, Anthropic, Bedrock, and Mistral, routing intelligently to balance performance and cost, while 300+ connectors ground every agent in live enterprise data. LLMOps tooling for tracing, evaluation, and automated QA lets teams catch failures before they reach production, making SimplAI a full-stack solution for industries like insurance, finance, and e-commerce where accuracy and compliance are non-negotiable.

• Vaethat is the AI-powered render enhancer that architects, 3D artists, and marketers use to upscale, denoise, and refine architectural renders without rerendering or manual editing. The platform's three presets — Precision, Detail, and Creativity — cover the full range of post-production intent in a single click, while its archviz-trained AI engine ensures geometry, lighting, and material choices survive the enhancement process intact. Batch project processing and a simple upload-and-export flow make Vaethat a direct replacement for the time-intensive Photoshop work or costly render farm reruns that typically end a visualization project.

  ---

• Humanoid: HMND SERIES E02 | Our Path to Global Leadership in Industrial AI Humanoid Robotics

• Unitree Robotics: A Whole Bunch of Robots Sending New Year Greetings to Everyone!

• Boston Dynamics: ST Engineering MRAS | Boston Dynamics

Welcome to this week's edition of Rushing Robotics! This edition showcases robots with insect vision, brain chips tested in space, and medical breakthroughs that stop bleeding in seconds. From AI moving to your desktop to humanoid robots joining factory floors, these innovations prove we're solving yesterday's impossible problems today.

🤯 Mind-Blowing

A 1,000-lens compound eye gives robots 180-degree vision and smell detection simultaneously, while brain chips successfully operated in orbit for the first time. A $250 desktop AI computer delivers 24 trillion operations per second without cloud dependency, and autonomous vehicles gained superhuman reflexes responding four times faster than human brains. Meanwhile, iron catalysts achieved 85% hydrogen fuel efficiency, potentially slashing fuel-cell vehicle costs by replacing expensive platinum.

🔊 Industry Insights & Updates

Light-powered computing is solving optimization problems faster than conventional systems using laser pulses at room temperature. Meta broke ground on a $10 billion AI data center in Indiana with 1 gigawatt capacity, while Apptronik secured $520 million to deploy Apollo humanoid robots already working at Mercedes-Benz and logistics facilities. Italian shipyards welcomed humanoid robot welders through a partnership targeting deployment within two years.

🧬 BioTech

MIT's magnetic mixer solved 3D bioprinting's cell-settling problem, enabling consistent functional tissue production. RNA micelles nearly eliminated metastatic cancer tumors in 26 days by combining chemotherapy with gene therapy. A revolutionary spray stops severe bleeding in under one second by transforming blood into gel, awaiting clinical trials for emergency deployment.

💡 Products/Tools of the Week

Parano.ai automatically tracks competitor activities across marketing, pricing, and product changes, delivering AI-summarized alerts to save research time. Vicoa enables seamless AI coding sessions across devices with real-time synchronization and one-tap approvals. QVeris AI provides a unified API connecting LLMs to 10,000+ third-party tools with sub-50ms latency and enterprise security. Google DeepMind's Project Genie transforms text or image prompts into interactive, photorealistic 3D environments in real-time, eliminating traditional modeling expertise requirements.

🎥 Video Section

Boston Dynamics showcases Atlas performing airborne maneuvers, Nick Builds constructs a laundry-folding robot in 24 hours, and UBTECH's autonomous logistics vehicle completes validation testing at Foxconn.

The convergence of biology, AI, and robotics is accelerating beyond prediction. These aren't concepts—they're working systems reshaping our world right now. What once seemed impossible is now deployed and operational, transforming industries and saving lives. Stay hungry, stay futurish!

🤯 Mind-Blowing

• A 1,000-lens sensor eye has been developed that gives robots 180-degree vision and the ability to detect smells simultaneously. Researchers at the Chinese Academy of Sciences created an artificial compound eye modeled after the fruit fly, fitting 1,027 visual units into a sensor measuring just 1.5 millimeters using advanced laser two-photon polymer printing. The device includes tiny hair-like structures called setae between the lenses that prevent moisture accumulation and protect from dust and debris. Engineers integrated a bionic nose using an inkjet-printed chemical array that changes color when detecting harmful gases, merging vision and smell into a single lightweight unit for small robotic systems. The bio-CE system successfully identified moving objects and navigated around obstacles during testing, detecting threats from both front and sides simultaneously without requiring head rotation.

• A wireless brain-computer chip has been tested in space for the first time by Chinese researchers, successfully operating in orbit to collect brain signals despite harsh space conditions. Northwestern Polytechnical University (NPU) developed and launched the implantable brain-computer interface device into orbit last December as part of a specialized space experiment platform. The operational BCI was immersed in simulated body-fluid settings during testing and successfully maintained consistent electroencephalogram collection despite extreme challenges, demonstrating that sensitive neural electronics can remain functional without degradation or short-circuiting outside Earth's atmosphere. The experiment provided crucial information about hardware resilience, noise interference effects, electrode durability in space environments, and how microgravity influences neural firing patterns. The NPU team created a flexible electrode array designed to conform to the brain's natural shapes ensuring a comfortable noninvasive fit, surpassing standard metal counterparts with dramatically improved signal stability metrics in animal testing.

• A pyramid-shaped desktop computer has been launched that runs AI tasks locally rather than on the cloud, offering 24 trillion operations per second at a price of $250. M5Stack introduced the AI Pyramid Pro, an edge-AI device equipped with 8GB RAM, 32GB internal storage, HDMI 2.0, four USB 3.0 ports, and two Type-C ports designed for real-time computer vision, speech comprehension, and object recognition. The device uses an Arm CPU and dedicated neural processor instead of a GPU to maintain energy efficiency and low power consumption while running small to medium-sized language models and managing multiple camera feeds simultaneously. Local AI processing provides enhanced privacy, reduced latency with near-instantaneous responses, elimination of per-use cloud fees, and the ability to operate without continuous internet connection. The Pyramid Pro targets edge-AI developers, embedded systems engineers, and creators working on advanced AI projects rather than general consumers.

• A new autonomous car vision system has been developed that responds to threats four times faster than the human brain. Researchers from China, the UK, Hong Kong, Saudi Arabia, and the US created a hardware-based reflex mechanism designed to accelerate decision-making in automated driving. The system uses a two-dimensional synaptic transistor array functioning as a motion detection chip that identifies changes in images in just 100 microseconds, significantly outpacing human perception which requires roughly 150 milliseconds to react. The technology employs a filter-then-process methodology inspired by human visual processing, disregarding entire images and recording only moving objects before relaying selected signals to computer vision algorithms.

• A new catalyst has been developed that achieves 85% hydrogen fuel efficiency by solving the decades-old problem of stabilizing iron catalysts for fuel cells. Scientists at Washington University in St. Louis created a technique using chemical vapor of gases to stabilize iron catalysts during thermal activation, making them viable for proton membrane fuel cells. This breakthrough could dramatically reduce fuel-cell vehicle costs, as platinum catalysts currently account for about 45% of a fuel cell stack's expense. Iron being plentiful and inexpensive presents a viable alternative to scarce platinum, though it previously struggled with stability in the acidic conditions required. The technology targets heavy-duty vehicles like transport trucks and buses that operate from centralized locations, making hydrogen refueling logistics more manageable.

🔊 Industry Insights & Updates

• A new light‑powered machine has been built that solves complex optimisation problems faster than conventional computers, using pulses of laser light to emulate magnetic spins in an Ising model. Researchers at Queen’s University, led by associate professor Bhavin Shastri, assembled a room‑temperature optoelectronic oscillator system that can process up to 256 coupled “spins” with just five optical components, and already outperforms older optical Ising machines by a wide margin. The device, demonstrates how billions of operations per second can be achieved without the need for cryogenic cooling, offering a promising pathway for tackling real‑world tasks such as drug discovery, cryptographic analysis, and logistics optimisation.

• Construction has started on a $10 billion AI data center in Lebanon, Indiana, by Meta that will provide 1 gigawatt of power capacity. The facility represents one of Meta's most substantial infrastructure investments and will serve as the company's second data center in Indiana, engineered to manage digital platforms and rapidly expanding AI workloads. The project is anticipated to generate over 4,000 jobs during peak construction phases with approximately 300 permanent positions once operational. Meta committed to matching electricity consumption entirely with clean energy, achieving LEED Gold certification, utilizing closed-loop liquid-cooling systems, and returning 100 percent of water consumed back to local watersheds through collaboration with Arable on irrigation technology for farmers.

• A $520 million funding round has been secured by US-based robotics company Apptronik to deploy its Apollo humanoid robots in manufacturing settings, achieving a $5 billion valuation with backing from Google and Mercedes-Benz. Apollo robots are already operational at partner facilities including Mercedes-Benz, GXO Logistics, and others where they assist in optimizing manufacturing and warehouse tasks. The flagship Apollo robot stands 5'8" tall, weighs 160 pounds, can handle loads of up to 55 pounds, and functions for four hours on a single battery charge with hot-swappable battery packs for continuous operation. Google DeepMind recently demonstrated Apollo's capabilities powered by Gemini AI, showcasing its ability to manage unfamiliar items, follow spoken instructions, and adjust in real-time to varying circumstances without retraining in new environments.

• A humanoid robot welder has been introduced to work alongside humans in Italian shipyards through a partnership between Generative Bionics and Fincantieri. The robot will function as a welder enhancing safety, boosting efficiency, and improving production quality in naval production facilities. Generative Bionics will develop the robot with artificial intelligence and advanced capabilities in manipulation, perception, and vision to oversee welding seams and operate in intricate environments. The four-year collaboration targets swift development with initial testing slated for completion by the end of 2026, with operational capabilities expected within the first two years followed by broader implementation at Fincantieri's Monfalcone shipyard.

Subscribe now

🧬 BioTech

• A magnetic mixer called MagMix has been developed by MIT researchers that creates the most uniform 3D-printed tissues by preventing cell settling during the printing process. The device addresses the problem where living cells denser than hydrogels sink to the bottom of printer syringes, causing blockages and uneven tissue distribution. MagMix consists of a small magnetic propeller inside the bioprinter syringe and an external motor that controls the propeller without direct contact with bio-ink, preventing contamination. Testing showed MagMix prevented cell settling for over 45 minutes of continuous printing while maintaining high cell viability, enabling production of functional muscle tissues. Ritu Raman, Eugene Bell Development Professor at MIT, stated the system allows mixing speeds to be fine-tuned for different bio-inks while exerting minimal stress on cells.

• Metastatic colorectal cancer tumors in mouse lungs were almost completely depleted within 26 days using newly developed RNA micelles that deliver chemotherapy and gene therapy directly to cancer cells. Researchers at The Ohio State University created these tiny molecular clusters by combining the chemotherapy drug gemcitabine with small interfering RNA that silences survivin, a protein helping cancer cells survive, and attached a ligand molecule to enhance binding to cancer cell receptors. The nanoparticles spontaneously target tumors without causing toxicity or immune responses. The study, showed the micelles attacked cancer in two ways by accumulating in tumor blood vessels and entering cells with the ligand's help. Co-author Daniel Binzel, a research assistant professor at Ohio State, explained that the chemotherapy kills cells while the small interfering RNA blocks survival gene expression, eliminating cancer through multiple pathways simultaneously.

• A revolutionary spray has been developed that stops severe bleeding almost instantly by transforming blood into a gel-like substance in under one second. Researchers at Korea Advanced Institute of Science and Technology (KAIST) created this hemostatic agent designed for military and emergency medical use where traditional methods prove ineffective. The spray contains three natural ingredients: alginate from seaweed that forms gel on contact with blood, gellan gum from bacteria that stabilizes the gel structure, and chitosan from crustacean shells that attracts blood cells to accelerate clotting. The powder absorbs approximately seven times its weight in blood and works without requiring direct pressure on wounds. The technology awaits clinical trials and regulatory approval before deployment in military medical kits and hospitals.

• ---

💡Products/tools of the week

• A new AI-powered competitive intelligence platform is now available to help businesses automatically track competitor activities and reclaim valuable research time. Parano.ai continuously monitors competitors' marketing, product, pricing, reviews, hires, funding, and website changes, detecting meaningful shifts with visual diffs and delivering AI-summarized alerts directly to email, Slack, or Microsoft Teams. The platform's AI analyzes patterns and context to explain why changes matter, enabling GTM, product, and leadership teams to act faster on strategic moves without wasting hours on manual research. Companies can set up tracking for competitor sets within minutes and receive comprehensive weekly summaries with executive insights, all without needing a credit card to start.

• A new cross-device AI coding companion has been launched to enable seamless remote development workflows. Remote AI coding sessions can now be managed from any device through Vicoa, a platform that runs AI agents like Claude Code and Codex remotely and provides a clean interface with real-time session synchronization, searchable history, push notifications, and one-tap code diffs and approvals. Developers can start a coding session on a laptop and seamlessly continue on a phone or tablet without losing context, while Vicoa handles session management, approvals, and secure connections. The tool installs as a simple CLI command and keeps AI central to the workflow, allowing agents to write, review, and interact with code while developers avoid babysitting long-running tasks. This makes Vicoa ideal for developers who want to maintain productivity on the go and keep their AI-driven workflows synced across all devices.

• QVeris AI is a Tool OS for AI agents that provides a single, OpenAI‑SDK‑compatible API and SDKs to discover, route, sandbox, and execute 10,000+ third‑party tools (search, finance, OCR, weather, etc.) so LLMs can automatically find, select, and call real‑world services; it includes plugins (Cursor), a QVerisBot assistant option, and examples to get started quickly. It’s built for production with distributed routing, smart failover, edge nodes (<50ms avg latency), 99.9% SLA, and enterprise security (encryption and fine‑grained access control). Developers and teams use it to eliminate adapter work, accelerate agent development.

• A new experimental research prototype has emerged that transforms simple text or image prompts into photorealistic, interactive 3D environments generated in real time. Google DeepMind developed Project Genie using Genie 3, a general-purpose AI world model that enables users to sketch worlds, build characters, and immediately explore navigable scenes that dynamically expand around them. The system simulates physics, remembers prior interactions, and allows creators, researchers, and prototypers to rapidly test game levels, storytelling scenes, and robotics or simulation concepts without traditional 3D modeling expertise. By emphasizing AI-driven on-the-fly world generation and remixing, this tool removes technical barriers and accelerates creative iteration for diverse applications across entertainment and research.

  ---

• Boston Dynamics and RAI Institute: Atlas Airborne | Boston Dynamics & ‪@rai-inst‬

• Nick Builds: I Made A Laundry Folding Robot In 24 Hours

• UBTECH Robotics: UBTECH Autonomous Logistics Vehicle Chitu α Completes PVT at Foxconn

Welcome to this week's edition of Rushing Robotics! Another week, another leap toward futures that once seemed impossible. We're witnessing the collision of biology and technology at unprecedented speeds—from cellular clocks being rewound to smart materials that morph like living skin. This edition captures the breakthrough moments where science fiction becomes medical fact, where quantum computing moves from theory to tangible hardware, and where AI agents evolve from simple chatbots into autonomous collaborators reshaping how we work.

🤯 Mind-Blowing

Get ready for jaw-droppers: FDA greenlights the first human trial for age reversal via cellular reprogramming in glaucoma patients, octopus-inspired 4D smart skin that camouflages and encrypts info, nanoparticle breakthroughs for safer CRISPR delivery, a bee-brain chip slashing navigation power use by 99%, and OpenAI's Codex app orchestrating AI agents for full dev lifecycles.

🔊 Industry Insights & Updates

Taiwan unveiled its first homegrown 20-qubit quantum computer, built entirely domestically for research and training. Perovskite solar cells hit 26.74% efficiency using molecular "shields" that neutralize light degradation, surviving 1,000 hours at 95% capacity. Chinese scientists demonstrated supercooling tech that drops liquids 30°C in 30 seconds using pressurized ammonium—potential relief for power-hungry AI data centers. The UAE partnered with Colossal Biosciences on a nine-figure BioVault to preserve 10,000+ species' DNA, open-sourcing genomes to combat projected 2050 extinction rates.

🧬 BioTech

Immunotherapy reduced arterial plaque in mice by targeting dysfunctional smooth muscle cells, with a matching imaging tracer spotting human coronary plaques in PET scans. Complete pancreatic tumor regression occurred in mouse models using triple-combination therapy blocking multiple KRAS signaling pathways simultaneously. Caltech's "bubble bots"—microscopic protein bubbles with enzyme engines—autonomously navigated to bladder tumors using chemical cues, reducing tumor weight 60% when delivering drugs via ultrasound-triggered bursting.

💡 Products/Tools of the Week

NodeTool offers open-source visual AI workflow building without code. Xpoz MCP lets AI assistants query live social media via natural language without API keys. Springhub converts prompts into automated workflows across 1,000+ integrations. Skills.sh serves as a package manager for modular Agent Skills extending AI agents with procedural knowledge.

🎥 Video Section

South China Morning Post presents Shanghai's first biomimetic AI robot. Westwood Robotics shows THEMIS Gen2.5 in continuous production. Humanoid unveils the KinetIQ AI framework. CnEVPost features Bolt, the fastest humanoid from MirrorMe.

The future is not approaching—it is being printed, programmed, and proven weekly. Each breakthrough collapses timelines and dissolves boundaries, promising a decade that will make the last century look like mere preamble. Stay hungry, stay futurish!

🤯 Mind-Blowing

• FDA approval has been granted for the first human trial of age reversal technology using cellular reprogramming, with Life Biosciences set to test the treatment on glaucoma patients shortly. David Sinclair's Boston startup will use viruses to deliver three reprogramming genes into patients' eyes, controlled by doxycycline antibiotic that activates the genes for approximately two months during monitoring. The approach aims to reset epigenetic switches on genes to restore cells to healthier states, based on the discovery that Yamanaka factors can turn cells back into stem cells like those in early embryos. The trial will initially involve about a dozen patients with glaucoma, a condition where high pressure damages the optic nerve, with the treatment injected into one eye of each participant. Sinclair claims that gradual loss of epigenetic information is the ultimate cause of aging and that partial reprogramming can reverse this process, though the technology carries risks including potential tumor formation when genes are activated in living animals. Michael Ringel described it as a starting bell for a new era of age reversal.

• Octopus-inspired programmable smart skin has been developed by Penn State researchers led by Hongtao Sun using 4D-printed hydrogel that dynamically changes appearance, texture, and shape in response to environmental stimuli. The halftone-encoded printing method programs binary patterns controlling material responses to heat, solvents, or mechanical stress, with demonstrations including Mona Lisa image encryption that becomes visible under specific conditions. Doctoral candidate Haoqing Yang explained the material enables adaptive camouflage and information hiding with security layers through mechanical deformation analysis. The smart skin transforms into complex 3D shapes within single layers.

• Was reported that new research is tackling the long-standing challenge of delivering CRISPR-Cas9 safely and efficiently into cells, aiming to move gene editing closer to real-world medical use. This work describes how scientists are developing DNA-free nanocarriers and other nanoparticle-based systems to reduce off-target effects and immune reactions while still achieving precise genome modifications. The article explains that by avoiding viral vectors and instead engineering smart nanomaterials, researchers hope to improve control over where and when CRISPR is active in the body. According to the piece, these advances could open doors for future treatments of genetic diseases, cancers, and other hard-to-treat conditions once safety and targeting issues are better managed. The report also emphasizes that while progress is significant, extensive testing and regulatory scrutiny will still be needed before such nanotechnology-enabled CRISPR therapies reach clinical routine.

• A groundbreaking computer chip inspired by bee brain navigation has been developed by European researchers from the InsectNeuroNano initiative. The prototype replicates how bees navigate using sky polarization and speed detection, achieving unprecedented energy efficiency by consuming less than one hundredth of a watt compared to conventional navigation chips that use over 7 watts. The specialized chip uses nanophotonic circuits that guide light through structures only billionths of a meter across, enabling ultra-compact positioning systems. Professor Elisabetta Chicca from the University of Groningen built virtual models of the chips based on insights from biologists, creating collaborative feedback between computer modeling and biological understanding. The research team envisions applications ranging from insect-sized robots for environmental cleanup and artificial pollination to low-cost environmental sensors, though Mikkelsen estimates real-world implementation will take approximately 10 years.

• OpenAI has launched the Codex app for macOS enabling developers to manage multiple AI agents running parallel tasks across software development lifecycles, with agents working in separate project-organized threads and worktree support preventing repository conflicts. Codex has evolved beyond code generation through skills that bundle instructions and scripts for tasks like implementing Figma designs, managing Linear projects, deploying to cloud platforms, and generating images, demonstrated by independently building a racing game using over 7 million tokens from one prompt. The app features Automations for scheduled background work on repetitive tasks with results in review queues, and includes system-level sandboxing that limits agents to editing files in their working folder while requiring permission for elevated commands. More than a million developers have used Codex since its April 2025 launch, with usage doubling since GPT-5.2-Codex launched in mid-December.

🔊 Industry Insights & Updates

• Was unveiled that Academia Sinica introduced Taiwan’s first homegrown 20-qubit superconducting quantum computer, marking a major milestone for the nation’s quantum technology ambitions. The report states that the research team built the quantum processor, control systems, and related software domestically, demonstrating the island’s capability to develop advanced quantum hardware rather than relying solely on foreign platforms. It explains that this 20-qubit machine is intended primarily for research use, enabling local scientists to experiment with quantum algorithms, error mitigation, and materials while training a new generation of quantum engineers. The article adds that officials emphasized the system as a foundation toward more powerful future devices and as a strategic asset for Taiwan’s competitiveness in next-generation computing.

• Researchers from China, Macau, and France have developed perovskite solar cells achieving 26.74 percent certified efficiency and maintaining over 95 percent initial efficiency after 1,000 hours of continuous light exposure using hindered amine light stabilizers. The team incorporated these stabilizers into the perovskite material to neutralize superoxide radicals that cause light-induced degradation, with the hindered amine transforming into a nitroxyl radical that operates in a regenerative cycle. The stabilizer also attaches to defects at grain boundaries, reducing trap sites and improving charge carrier movement while enabling larger crystal grains and smoother films. Devices were produced under ambient air conditions, demonstrating practical manufacturability, with the researchers stating that light instability can be addressed chemically rather than being an unavoidable issue.

• A breakthrough cooling technology cooled a liquid from room temperature to sub-zero levels in less than half a minute, as Chinese scientists demonstrated in recent experiments. The system, powered by ammonium thiocyanate in water under pressure, mimics squeezing a wet sponge to trigger rapid heat absorption upon pressure release. In tests, it dropped temperatures by 30 degrees Celsius at room temperature and over 50 degrees in hotter conditions, offering a solution for energy-intensive AI data centres in China and the US. Zhang Tong reported on this innovation from Beijing for the South China Morning Post, highlighting its potential to manage soaring cooling demands in the AI race.

• The UAE has partnered with Colossal Biosciences in a nine-figure initiative to protect global biodiversity through pioneering conservation and de-extinction research. The facility will preserve genetic material from more than 10,000 species, initially focusing on the 100 most imperiled species not currently preserved elsewhere, using automated robotics and AI-powered monitoring systems. Majed Al Mansoori, Executive Director of Museum of the Future, stated the first year will prioritize fieldwork and DNA research to lay scientific groundwork for future biodiversity protection. Ben Lamm, Co-Founder and CEO of Colossal Biosciences, emphasized the urgent need for a distributed network of BioVaults as nearly half of Earth's species could face extinction by 2050. The BioVault will create reference genomes, share non-proprietary data openly with global scientists, and allow visitors to observe live research.

Subscribe now

🧬 BioTech

• An antibody-based immunotherapy has reduced arterial plaque in mice by targeting modulated smooth muscle cells that drive inflammation and dangerous plaque formation, according to Washington University School of Medicine researchers led by Kory Lavine. The bispecific T cell engager molecule enlists the immune system to eliminate these dysfunctional cells, identified through single-cell profiling of over 150,000 cells from 27 human coronary arteries using fibroblast activation protein as a surface marker. The therapy reduced plaque amounts, diminished inflammation, and improved plaque stability in mouse models compared to untreated animals. The team also developed an imaging tracer that successfully identified coronary plaques in PET/CT scans of patients with coronary artery disease,

• Complete tumour regression in pancreatic cancer has been achieved in mouse models using a triple combination therapy targeting KRAS-driven signalling networks, according to a study by Mariano Barbacid's research team published in the Proceedings of the National Academy of Sciences. The research team employed a multi-node blockade strategy designed to inhibit multiple signalling pathways in genetically engineered mouse models of pancreatic ductal adenocarcinoma, demonstrating complete regression of established tumours with durable responses and no evident drug resistance. The approach was well-tolerated in mice and represents a significant departure from single-agent targeting that has historically failed in pancreatic cancer due to the disease's dense tumour microenvironment, early metastasis, and challenging molecular drivers like KRAS.

• Simple biocompatible microrobots called "bubble bots" have been developed by a Caltech-led team to autonomously target and treat tumors, representing a major advancement in medical microrobotics. The team, led by Wei Gao, professor of medical engineering at Caltech and Heritage Medical Research Institute Investigator, created thousands of microbubbles with protein shells using ultrasound to agitate a bovine serum albumin solution, then attached the enzyme urease to act as a tiny engine that propels the robots using urea as biofuel. The researchers developed two versions: one with magnetic nanoparticles for external steering via magnets and ultrasound imaging, and a second intelligent version with catalase enzyme that automatically moves toward tumors by detecting high hydrogen peroxide concentrations through chemotactic behavior. Once the bubble bots reach their target, focused ultrasound bursts the bubbles to release anti-cancer drugs like doxorubicin, with the bursting action enhancing drug penetration into tumors. When scientists injected mice with bubble bots to deliver anti-tumor therapeutics, they observed roughly 60 percent decrease in bladder tumor weight over 21 days compared to mice given the drug alone.

  ---

💡Products/tools of the week

• A new open-source visual workflow builder has been launched that enables users to design and execute AI pipelines through drag-and-drop nodes without writing code. NodeTool operates locally and connects various AI capabilities including large language models, multimodal content generators for images, audio and video, retrieval-augmented generation with vector search, and autonomous agents. The platform supports both local models and cloud APIs, offering streaming execution with intermediate outputs for debugging transparency. Users can rapidly prototype and deploy complex AI workflows for media generation, document indexing, and agent automation while maintaining data privacy and avoiding vendor lock-in through integration with multiple providers including Hugging Face, llama.cpp, OpenAI, and Replicate.

• A new MCP server called Xpoz MCP has been launched that allows AI assistants including Claude and ChatGPT to query and analyze live social media platforms such as X/Twitter, Instagram, TikTok and Reddit using natural-language prompts. The server returns structured posts, profiles, engagement metrics and historical data without API keys or rate limits. By exposing a streaming, MCP-compliant connector featuring advanced filters, smart caching and OAuth2 authentication, it converts large language models into real-time social-intelligence agents capable of trend tracking, viral content analysis, influence mapping and threat detection. The solution makes conversational social listening and automated agent workflows fast, cost-effective and straightforward to integrate into research and marketing operations.

• An AI automation platform has been launched that converts prompts into reusable, no-code workflows called Recipes and scheduled agents operating continuously across more than 1,000 integrations. Springhub combines over 350 text and image models including GPT-5.2, Claude Opus and Gemini, allowing users to select the optimal model for each specific task. The platform features a visual canvas, conditional routing, data flows between steps, grounded answers from uploaded knowledge bases, and agentic actions such as sending emails, creating issues or posting to Slack. Users can automate multi-step tasks, eliminate repetitive prompts, and deploy AI that executes work autonomously.

• A registry and package manager for Agent Skills has been launched, providing modular, sharable packages that extend AI agents like Anthropic's Claude with procedural knowledge, scripts, templates and invocation controls. Skills.sh enables developers and teams to discover, install skills using commands like npx skills add owner/repo, and run them by loading concise metadata into an agent's context while providing on-demand full instructions and executable resources. This approach specializes general large language models for reliable, repeatable workflows without bloating the model context, allowing users to transform generic AI into task-specific agents and share domain expertise across projects.

  ---

• South China Morning Post: World’s first ‘biomimetic AI robot’ debuts in Shanghai

• Westwood Robotics: THEMIS Gen2.5 + AOS: Keep Moving, Keep Producing

• Humanoid: Introducing KinetIQ | Humanoid’s AI framework

• CnEVPost: Bolt — world’s fastest humanoid robot from Chinese firm MirrorMe

Welcome to this week's edition of Rushing Robotics! Every week the boundary between "emerging technology" and "working infrastructure" gets a little blurrier. This week brought AI that scores like a creative human, robots running cross-platform apps like smartphones, and gene therapies you can apply like lotion. We're not talking about future promises anymore—we're tracking deployments, benchmarks, and real products hitting real users. Here's what landed this week.

🤯 Mind-Blowing

This week’s breakthroughs blur the line between imagination and implementation. AI systems are now matching or beating average human scores on creativity benchmarks—yet still shine brightest when paired with human guidance. Robots are getting their own cross‑platform “app store,” turning hardware into a software‑defined ecosystem. Materials science and 3D printing are quietly reinventing what’s physically possible, while NVIDIA’s video‑world models let robots plan over long horizons using pure vision. Put together, it’s a picture of tools that don’t just execute instructions, but help us explore entirely new design spaces.

🔊 Industry Insights & Updates

On the industry side, the center of gravity keeps shifting from cars and chips to robots and inference. Tesla is sunsetting its legacy flagships to free capacity for Optimus humanoids—an explicit bet that the next “vehicle” of value is robotic labor, not EVs. Microsoft’s Maia 200 shows how seriously hyperscalers take AI economics, squeezing more tokens per watt and dollar to feed ever-larger models. NASA’s new Athena supercomputer slots AI right alongside rockets and aircraft as core infrastructure, while Hyundai’s humanoid trials reveal how quickly factories may move from fixed automation to adaptable robotic co‑workers.

🧬 BioTech

Biotech is quietly having its own AI‑like moment. A cancer‑linked molecule that clears misfolded brain proteins hints at drug strategies that could decouple Alzheimer’s risk from aging. Meanwhile, CRISPR is stepping out of the lab and onto the skin, with topical gene editing that actually restores function in inherited skin disorders. And by targeting the right exon in a single resistance gene, researchers are reopening treatment paths for head and neck cancers once thought out of options. The pattern: more precise edits, closer to root causes, with delivery methods that feel less like sci‑fi and more like skincare.

💡 Products/Tools of the Week

The tools arriving this week don't just automate—they compress entire workflows into seconds. Mowgli turns napkin sketches into production specs and themed UI, letting designers skip straight to iteration. ModelMonkey brings Claude into your spreadsheet to build financial models, pull live data, and explain every formula it writes. ArchRender bypasses complex 3D lighting setups, rendering photorealistic architectural visuals from sketches or models in moments. And HouseGPTs lets homeowners and agents visualize redesigns and staging instantly, cutting weeks off renovation planning.

🎥 Video Section

If you want to see the future instead of just reading about it, this week’s videos are a good start. Figure’s Helix 02, LimX Dynamics’ autonomous deployments, and Fauna’s real‑world humanoid platform all show different answers to the same question: what does a general‑purpose robot actually look like when it leaves the lab? Watch the gaits, the recovery, the manipulation—then imagine what happens when software updates hit at internet speed.

We’re still early in all of this, which is exactly why it’s so exciting. Creativity benchmarks remind us humans still dominate the top tier, even as AI stretches what’s possible for everyone else. Robots are moving from demos to deployment, and gene editing is edging toward everyday therapies.. Stay hungry, stay futurish!

🤯 Mind-Blowing

• Leading AI models have demonstrated creativity levels surpassing average humans in a landmark study comparing 100,000 participants with generative AI systems. Researchers at Université de Montréal employed the Divergent Association Task, a psychology tool measuring creativity by evaluating how unrelated generated words are from each other. The team tested ChatGPT, Claude, and Gemini against human participants, finding that while AI exceeded many humans in divergent creativity scores, half of the participants performed better than the machines, with the top 10% significantly outscoring AI. Professor Karim Jerbi noted that the study highlights how AI profoundly transforms how creators imagine and explore rather than replacing them. Published in Scientific Reports, the research also examined creative writing tasks, revealing that AI expressed maximum creativity when guided by humans, suggesting technology's role as an assistive tool in creative endeavors.

• A robot app store went live on Apple's App Store on January 27, 2026, when OpenMind launched its cross-platform marketplace. The platform brings together eight competing companies including UBTECH, Zhiyuan Robotics, and Fourier, signaling the industry's shift from hardware competition to software ecosystem development. OpenMind's OM1 operating system enables developers to write code that works across different robot types, from bipedal humanoids to four-legged robotic dogs. The system uses capability-oriented programming where developers issue commands to abstract motion capabilities rather than specific joints. Security sandboxes prevent direct access to motor drivers, with all motion commands passing strict feasibility checks. The platform includes a web-based simulator and crowdsourced testing network using real robots. OpenMind introduced a skills economy with encrypted execution environments for protection and dynamic pricing based on performance metrics.

• A method to transform MXenes into one-dimensional forms that could enhance batteries, biosensors, and wearable technology has been developed by scientists at Drexel University. The scalable technique rolls flat two-dimensional MXene sheets into tubular structures called MXene nanoscrolls, approximately 100 times slimmer than human hair with improved electrical conductivity. Yury Gogotsi explained that one-dimensional forms provide distinct advantages in scenarios requiring rapid transport or mechanical support. Cheng Zhang noted that transitioning from 2D nanosheets to 1D scrolls mitigates nano-confinement issues that challenge ion movement. The process uses water to adjust surface chemistry, creating structural imbalance that causes layers to separate and curl into compact scrolls. The team applied this technique to six distinct MXenes producing grams of nanoscrolls with controlled shapes. The tubular design increases active surface area, enhancing accessibility for ions and molecules crucial for battery performance and chemical sensing.

• A novel 3D printing technique named CRAFT has been introduced by researchers from The University of Texas at Austin in collaboration with Sandia National Laboratories, enabling production of objects with diverse mechanical and optical characteristics using a single cost-effective material. The team succeeded in creating a lifelike replica of a human hand by transforming an affordable liquid into a versatile material. Zak Page remarked that they control molecular order in three-dimensional space, completely changing the mechanical and optical properties of a material. CRAFT converts cyclooctene into solid structures through projection of different light patterns, with researchers adjusting light intensity to influence molecular order as material hardens. The team can create parts that are hard and transparent while keeping adjacent sections soft and opaque within a single component. CRAFT presents a superior substitute for medical cadavers by replicating intricate relationships of bone, ligament, and muscle within a single model. Beyond medical applications, CRAFT can produce bioinspired materials for energy absorption in helmets and armor as well as soundproofing.

• An innovative method for robot management has been unveiled by NVIDIA as Cosmos Policy, utilizing extensive video prediction models to enhance control and planning functions. NVIDIA refines a pre-existing video world model termed Cosmos Predict using data from robot demonstrations, which has already acquired understanding of how physical environments change over time. Robot actions, physical conditions, and task results are integrated into the model's internal temporal framework, enabling it to forecast subsequent actions and anticipated consequences. Benchmark evaluations show impressive success rates on multi-step tasks requiring long-term reasoning, matching or surpassing existing techniques while needing considerably fewer training demonstrations. The model can produce and assess various potential action sequences during inference, allowing robots to choose actions likely to succeed over extended periods. Hands-on experiments involving dual-arm manipulation successfully completed long-horizon tasks based solely on visual input.

🔊 Industry Insights & Updates

• Tesla will cease production of the Model S sedan and Model X SUV as CEO Elon Musk announced during a call with investors on Wednesday, signaling a shift away from traditional automotive programs towards artificial intelligence and robotics. This announcement coincided with the release of the company's latest quarterly earnings revealing a decline in vehicle sales but profits that exceeded expectations. Musk stated that he expects to wind down S and X production this quarter. The Model S debuted over ten years ago cementing the company's reputation in the U.S. automotive market, while the Model X was introduced with unique Falcon-wing doors and high-end appeal. The Fremont, California factory will be repurposed to produce the humanoid robot Optimus, with Musk's long-term vision to manufacture 1 million robots annually at the site

• A breakthrough inference accelerator from Microsoft called Maia 200 has been introduced to dramatically improve the economics of AI token generation. The chip, manufactured by Taiwan Semiconductor Manufacturing Co. using the 3nm process, packs 140 billion transistors and is tailored for large-scale AI workloads. Microsoft designed Maia 200 to deliver over 10 petaFLOPS in FP4 and over 5 petaFLOPS in FP8 performance within a 750W SoC power envelope, with its memory subsystem featuring 216GB HBM3e at 7 TB/s and 272MB of on-chip SRAM to keep massive models fed and highly utilized. The new accelerator achieves 30% better performance per dollar than the latest generation hardware in Microsoft's fleet and outperforms competing chips with three times the FP4 performance of Amazon's Trainium3 and FP8 performance above Google's TPU v7. Microsoft is deploying Maia 200 as part of its heterogeneous AI infrastructure to serve the latest GPT-5.2 models from OpenAI and power Microsoft Foundry and Microsoft 365 Copilot.

• NASA's most advanced supercomputer named Athena has been unveiled on January 27, intended to support upcoming initiatives in space exploration, aeronautics, and scientific inquiry. Athena stands as the latest flagship within NASA's High-End Computing initiative at Ames Research Center in Silicon Valley. Dan Murphy remarked that exploration has perpetually pushed NASA to the limits of computational capabilities. Athena provides processing power for simulating complex rocket launches and modeling fuel-efficient aircraft designs, enabling NASA to conserve millions by minimizing physical testing. The system acts as a training hub for large-scale artificial intelligence foundational models analyzing vast amounts of satellite and mission data. Athena achieves peak performance exceeding 20 petaflops with 1,024 nodes equipped with AMD EPYC processors and 786 TB of memory.

• Humanoid robot trials have begun at Hyundai Motor’s US manufacturing facilities, with proof-of-concept testing underway at the company’s Georgia plant since late last year to evaluate real-world functionality and gather operational data ahead of potential commercialization. The automaker announced plans in early January to implement a scalable platform capable of deploying up to 30,000 humanoid robots annually by 2028 to automate repetitive factory tasks. Hyundai intends to deploy Atlas, the humanoid robot developed by its robotics division Boston Dynamics, at the Georgia site starting in 2028, with initial responsibilities focused on parts sorting before advancing to more complex assembly tasks by approximately 2030. The production-ready Atlas robot, showcased at CES 2026 in Las Vegas on January 5, can lift weights up to 110 pounds and operate in industrial conditions with temperatures ranging from -4°F to 113°F, featuring 56 degrees of freedom to handle high-risk and repetitive activities. Hyundai is developing Software-Defined Factories that gather extensive data to allow robots to continuously refine their skills and is creating Robot Manufacturing Application Centers where robots will be trained in tasks like welding, turning, and assembling, in collaboration with NVIDIA for AI computing and simulation technologies.

Subscribe now

🧬 BioTech

• A specific molecule found in some cancer cells has been discovered to break down misfolded proteins in the brain, potentially explaining why cancer and Alzheimer's disease rarely occur together. University of Toronto researchers identified that certain cancers release a protein called cystatin-C (Cyst-C), which travels through the body and helps clear toxic amyloid clumps linked to Alzheimer's. The 15-year study used mouse models transplanted with human lung, prostate, and colon tumors, finding that mice with cancer remained free of cognitive decline plaques. Cyst-C binds to harmful amyloid clusters and activates the brain's immune cells, microglia, to break down the plaques. When researchers disabled the Cyst-C pathway, the protective effect vanished. The findings, published in Cell, open doors for developing drugs that degrade misfolded plaques without requiring cancer, and advance the search for existing cancer treatments beneficial for the brain.

• A topical CRISPR-based gene therapy has been developed that corrects genetic skin disorders through direct application to the skin. UBC Medicine researchers, working with the Berlin Institute of Health at Charité, created the first gene therapy capable of correcting faulty genes when applied directly to human skin, as detailed in Cell Stem Cell. Dr. Sarah Hedtrich, an associate professor at UBC's School of Biomedical Engineering and senior author, explained that the treatment corrects disease-causing mutations using a safe, scalable approach that addresses the root cause. The therapy demonstrated effectiveness in correcting the genetic mutation behind autosomal recessive congenital ichthyosis, a rare inherited disorder affecting one in 100,000 people. Testing showed the treatment can restore up to 30 percent of normal skin function. The treatment uses lipid nanoparticle technology to deliver gene-editing tools through microscopic openings created by laser, reaching skin stem cells beneath the surface.

• An important advance in treating head and neck cancers has been made through CRISPR gene editing that overcomes drug resistance in tumors that stopped responding to treatment. The ChristianaCare Gene Editing Institute team published their findings in Molecular Therapy Oncology, demonstrating a method to restore chemotherapy effectiveness by targeting the NRF2 gene rather than single proteins common in traditional drug development. Head and neck cancer is the seventh most common cancer worldwide with cases expected to rise 30 percent every year by 2030, and many patients reach a point where treatment no longer works. The researchers showed CRISPR can successfully disrupt NRF2 in head and neck cancer cells and esophageal cancer cells, building on earlier lung cancer studies where blocking NRF2 improved survival in animal models. Dr. Natalia Rivera-Torres emphasized breaking through the wall of drug resistance that so many patients face. The study revealed that the location of the CRISPR cut within the NRF2 gene makes a big difference, with targeting exon 4 reducing NRF2 levels by 90 percent compared to less effective exon 2 editing.

  ---

💡Products/tools of the week

• A new AI design copilot transforms napkin sketches into production-ready specifications and high-fidelity screens. Mowgli guides users through a structured questionnaire, automatically generating user journeys, data models, and themed UI elements. The platform supports natural-language iteration and exports work as Figma files, React+Tailwind code, or master prompts. Designed for designers, PMs, and founders, Mowgli delivers faster, more reliable design-to-development handoffs.

• An AI-powered add-on for Google Sheets and Excel has been introduced to automate financial modeling and spreadsheet analysis directly within workbooks. ModelMonkey uses Anthropic's Claude Sonnet 4.5 to build 3-statement models, DCF valuations, and cap tables through natural-language prompts, while pulling data from SEC filings, web sources, and data warehouses. The platform autonomously executes multi-step workflows, writes dynamic and auditable formulas into cells, and creates charts while applying firm-specific modeling conventions. ModelMonkey explains every change and formula it generates, providing one-click undo functionality so analysts and finance teams can review and trust edits without copying and pasting between separate tools. The add-on also debugs existing formulas, ensuring accuracy and maintaining workflow continuity for financial professionals.

• Photorealistic architectural visuals can now be generated in seconds from 3D models, sketches, or photos through a new AI-powered rendering platform. ArchRender uses cloud AI to interpret geometry, scale, and perspective, then applies materials and lighting while stylizing scenes based on natural-language prompts or uploaded moodboards. The platform enables designers and architects to bypass complex lighting and setup workflows, instead iterating rapidly with controls for seasons, times of day, edit modes, and design variants. ArchRender exports high-resolution upscaled renders suitable for client presentations, transforming the traditional rendering process that previously required extensive technical setup and rendering time.

• A new AI platform converts home photos into photorealistic interior and exterior redesigns within seconds. HouseGPTs applies generative models to propose style variations, 3D renderings, virtual staging, and furniture and color suggestions from a growing design library. Homeowners, designers, real estate professionals, and flippers visualize renovation ideas and stage properties instantly, cutting weeks of wait time and architect costs while accessing realistic, editable concepts.

  ---

• Figure: Introducing Helix 02

• LimX Dynamics: Oli Demonstrates the World’s First Scalable Autonomous Deployment

• Fauna Robotics: Meet Sprout | A Humanoid Robotics Platform for the Real World

Welcome to this week's edition of Rushing Robotics! Another week, another leap forward in artificial intelligence and emerging technologies. From molecules designed on demand to AI code that now powers nearly a third of all new software, we're watching intelligence reshape how we discover, build, and heal. This edition brings you breakthrough research that's shortening the path from lab bench to real-world impact, industry moves positioning AI for the physical world, biotech innovations that quite literally grow inside living cells, and the tools putting these capabilities into more hands every day.

🤯 Mind-Blowing

PropMolFlow from University of Florida and NYU generates molecular candidates 10 times faster than existing methods without sacrificing accuracy, producing chemically valid structures with correct bonding patterns over 90% of the time . AI-assisted coding now accounts for 29% of all newly written U.S. software functions—up from 5% in 2022—boosting productivity by 3.6% and adding $23-38 billion in annual economic value . Revoice, a washable wearable choker from Cambridge, uses throat sensors and AI to give stroke survivors their voice back by interpreting muscle vibrations in real time . A new analogue chip from Politecnico di Milano could cut computing energy by up to 5,000 times by processing calculations directly in memory . Integrated Quantum Technologies' AIQu™ VEIL™ platform lets businesses deploy AI without exposing sensitive data through dynamically computed anonymity .

🔊 Industry Insights & Updates

Microsoft's Rho-alpha interprets natural language commands for robots doing two-handed tasks, pushing AI into the physical world . The AM2PC project's 3D-printed cooling hit 600 watts—50% above target—while cutting data center energy use . MuWave secured funding to commercialize gyrotron tech for fusion power . TISQ integrates p-type and n-type semiconductors into single molecules for more stable organic solar cells .

🧬 BioTech

Slovenian researchers printed 3D micro-objects inside living cells using ultrafast lasers . Zurich clinicians are using denovoSkin lab-grown grafts to treat severe burns from the January 1 Crans-Montana fire that killed 40 . SPARK Microgravity announced Europe's first commercial orbital cancer lab to study tumors without gravity's interference.

💡 Products/Tools of the Week

AyeWatch scans a billion sources 24/7 with AI for real-time alerts . Gobii deploys AI workers for web automation without APIs . CanvasGPT combines research and full-stack code in one workspace . Janus animates avatars from webcam feeds for streaming .

🎥 Video Section

This week's video lineup shows humanoid robots making the leap from lab demos to real work environments. UBTECH showcases the first humanoid robotic worker deployed at SANY RE, while PNDbotics' Adam full-size humanoid demonstrates increasingly fluid movement with its Charleston Dance Version 2.0 . Skild AI reveals how their systems are learning complex manipulation tasks simply by watching human videos, and Agile Robots SE unveils Agile ONE—their next evolution in physical AI that promises more adaptive, capable robotic systems .

AI is moving from prediction to creation—building molecules, writing software, restoring voices, walking into factories. The pace is breathtaking and the best is still ahead. Stay hungry, stay futurish!

🤯 Mind-Blowing

• A new generative AI method developed by researchers at the University of Florida and New York University can produce molecular candidates approximately 10 times faster than current techniques without sacrificing accuracy or chemical validity, offering promise for accelerated drug and materials discovery. The PropMolFlow system inverts traditional scientific discovery by allowing scientists to specify desired properties and then find matching molecular structures, rather than discovering useful compounds by accident and later understanding why they work. According to Mingjie Liu, an assistant professor in the University of Florida's Department of Chemistry and one of the paper's authors published in Nature Computational Science, computational speed directly translates to discovery speed in this field, making the advance particularly meaningful. The innovative algorithm finds more direct paths from random noise to valid molecular structures, consistently outperforming baseline models on structural validity by generating molecules with correct bonding patterns and appropriate geometries more than 90 percent of the time.

• A new study published in Science reveals that AI-assisted coding has reached 29% of all newly written software functions in the United States by early 2025, up from just 5% in 2022, according to research from the Complexity Science Hub that analyzed over 30 million Python contributions from approximately 160,000 developers on GitHub. The research team led by Simone Daniotti used a specially trained AI model to identify AI-generated code through tools like ChatGPT or GitHub Copilot, finding that generative AI increased programmers' productivity by 3.6%, translating to between $23 and $38 billion in annual value for the U.S. economy. Frank Neffke, who leads the Transforming Economies group at CSH, noted that while the U.S. leads at 29%, Germany reaches 23%, France 24%, and India 20%, with China at 12%. Less experienced programmers use generative AI in 37% of their code compared to just 27% for experienced programmers, but the productivity gains are driven exclusively by experienced users

• Regained speech after stroke is the goal of Revoice, a comfortable, washable wearable choker that uses sensitive sensors and AI to interpret speech-related throat signals. The University of Cambridge team designed it for people with dysarthria, where weakened facial, oral, and throat muscles make speech difficult even when the person knows what they want to say. The system avoids invasive brain implants by capturing throat muscle vibrations and heart rate, then generating intended words and sentences in real time.

• Reduced by up to 5,000 times, energy consumption in computing could drop dramatically with a new analogue chip that executes calculations directly within memory arrays. Daniele Ielmini, a professor in the Department of Electronics, Information, and Bioengineering at Politecnico di Milano, led the development of the fully integrated analogue accelerator built using standard CMOS technology compatible with current silicon manufacturing. Testing revealed the chip achieved accuracy comparable to traditional digital systems but with reduced power consumption, lower computing latency, and a smaller silicon footprint.

• A newly introduced AI infrastructure platform enables businesses to implement artificial intelligence without exposing sensitive data, addressing persistent privacy and security challenges for organizations under stringent regulations. Integrated Quantum Technologies has unveiled its AIQu™ VEIL™ platform, which processes data in anonymized and compressed formats before integration into AI systems using a technique called Dynamically Computed Anonymity or DCA™ that transforms data into vectorized forms. Guib, CEO and Chairman of Integrated Quantum Technologies, stated that VEIL™ enables companies to harness AI without compromising their most critical data. Jeremy Samuelson, Executive Vice President of AI and Innovation who conceptualized VEIL™, noted that by processing exclusively mathematically anonymized data, they are creating effective and secure AI systems at scale.
  

🔊 Industry Insights & Updates

• Launched by Microsoft Research to extend robot capabilities beyond controlled factory settings, Rho-alpha is a new AI model designed to tackle a persistent robotics challenge: adapting to unpredictable, real-world conditions. The system represents Microsoft's first robotics-centric model from its Phi vision-language AI lineage and is positioned as part of a broader transition toward "physical AI," where intelligent agents engage directly with the tangible world rather than being limited to digital realms. Unlike conventional industrial robots that rely on inflexible task scripts, Rho-alpha interprets natural language commands and converts them into control signals for robots executing intricate two-handed manipulation tasks, with Microsoft currently assessing the system on dual-arm platforms and humanoid robots.

• A new passive two-phase cooling approach has been demonstrated to help data centers handle rising server and GPU heat loads while cutting the energy used for cooling. The AM2PC project, led by the Danish Technological Institute and Heatflow with partners from Belgium and Germany, developed a 3D-printed aluminum cooling component intended for data centers and HPC environments. In trials, the solution reached 600 watts of performance, exceeding the original 400-watt target by 50%.

• Backed by seed investment intended to accelerate deep-tech commercialization, MuWave will build and refine next-generation gyrotrons and related microwave products that can reduce risk and speed progress toward practical fusion power. Co-founder Helen Webster called the initial funding a milestone that validates the problem the team is solving and supports building momentum toward larger impact, while the UKI2S investment—managed by Future Planet Capital—highlighted the goal of turning scientific know-how into long-term industrial and economic outcomes beyond fusion alone. MuWave’s near-term plan includes producing and supplying high-powered microwave products within the UK while developing an export-ready technology base for global markets.

• Formed as a “shortcut” toward better organic thin-film solar cells, TISQ addresses a key bottleneck: the interface between p-type and n-type materials must be precisely tuned to separate and transport charges quickly after light absorption. Osaka Metropolitan University’s Takeshi Maeda said traditional fabrication often physically mixes p-type and n-type molecules, but small processing changes can cause inconsistent mixing, unstable structures, and reduced device performance. The team instead pursued a single-molecule strategy where p- and n-type semiconductor components are integrated into one molecule so nanoscale p/n heterojunctions can arise through molecular self-assembly.

Subscribe now

🧬 BioTech

• Printed 3D micro-objects were formed inside living cells, demonstrating a way to fabricate free-standing structures where cells normally cannot take up large solids. The Slovenia-based research team introduced 10–15 micrometer droplets of a commercial resist into HeLa cells and then scanned an ultrafast laser to harden the material only where needed, producing precise internal microstructures. The approach points toward new experiments that monitor cellular activity or test how cells respond to physical stimuli, while keeping the surrounding environment largely intact.

• Began after a catastrophic New Year’s fire in Crans-Montana, Switzerland, an advanced treatment effort is now using lab-grown skin grafts to help survivors with severe burns recover. The blaze at the Le Constellation bar on January 1, 2026 killed 40 people and injured 119, with an estimated 80–100 survivors suffering serious burns, many over more than 60% of their bodies. Clinicians in Zurich are using denovoSkin, a living skin graft grown from each patient’s own cells, to support wound closure and healing in the most severe cases.

• Announced at a public forum titled "Curing Cancer in Space," the orbital laboratory will enable experiments on three-dimensional tumor growth and cell signaling without gravitational interference. SPARK Microgravity executives explained that gravity influences cell growth, positioning, and interactions on Earth, creating limitations that are lifted in orbit where researchers can investigate intricate cancer behaviors more clearly. CEO and co-founder Allison Bajet illustrated the difference using an audio analogy, comparing Earth-based experiments to listening to a symphony in a construction zone while microgravity is like entering a soundproof room where every biological note becomes crystal clear.

  ---

💡Products/tools of the week

• New always-on monitoring is now available with AyeWatch, a 24/7 AI-powered tool from AyeSoul that uses its SOUL Watcher multi-agent AI to scan over a billion sources across news, blogs, social, research papers, and webpages. AyeSoul’s AyeWatch sends real-time, rule-based alerts only when relevant matches or significant changes occur, filtering duplicates and noise while learning from user feedback to improve relevance. By prioritizing actionable items, AyeWatch helps researchers, investors, journalists, founders, and other power users stay ahead without constant manual searching, supported by cross-platform apps and scalable free-to-paid plans.

• Always-on web automation is now easier with Gobii, a cloud platform that deploys AI-powered digital workers to run web-based tasks and continuous monitoring 24/7. Gobii’s agents can browse sites even without APIs, execute JavaScript, rotate proxies, extract and synthesize data, and take action on scheduled cadences or event triggers. With prebuilt worker templates for recruiting, sales, intel, and compliance plus integrations with Slack, Google Drive, CRMs, and APIs, Gobii helps teams offload repetitive research, enrichment, and operational workflows so AI handles data collection and basic decision-making while humans focus on higher-value work.

• Live product building is now centralized in CanvasGPT, an AI-powered infinite workspace that uses autonomous canvas agents like Patch plus integrated models such as Gemini, Claude Opus, Nano Banana Pro, ElevenLabs, and Kling to turn ideas into working products. CanvasGPT supports brainstorming, web research, high-fidelity image/audio/video generation, full-stack code production, and workflow automation directly on a persistent, shareable canvas. With AI embedded into core components like a built-in database, auth, and real-time sync, CanvasGPT helps non-developers and teams prototype, build, and deploy interactive apps without tool handoffs, while a credits-based system scales usage and model choice for faster creation and fewer silos.

• Real-time avatar animation is now available with Janus, an AI-powered character animation platform that turns a live webcam feed into animated avatars for streaming, video calls, and content creation. Janus uses AI facial-tracking and animation models to map expressions and movements onto pre-made or custom characters instantly, running in-browser or through a Chrome extension with no special hardware required. With broad streaming support, Janus enables live output to Discord, Zoom, Twitch, YouTube, OBS, and more while keeping setup lightweight for creators and teams.

  ---

• UBTECH Robotics: Meet The First Humanoid Robotic Worker at SANY RE!

• PNDbotics: Adam Full-Size Humanoid Robot Charleston Dance Version 2.0

• Skild AI: Learning by Watching Human Videos

• Agile Robots SE: The next evolution in Physical AI: Agile ONE

Welcome to this week's edition of Rushing Robotics! As we push further into 2026, the pace of innovation shows no signs of slowing. This week brings us extraordinary breakthroughs spanning robotics, quantum computing, bioengineering, and nuclear-powered AI infrastructure. From self-healing electronic nerves that give robots human-like reflexes to 3D-bioprinted livers destined for transplantation, the future is unfolding before our eyes. Let's dive into the developments that are reshaping our world.

🤯 Mind-Blowing

Robots are gaining graded “pain” reflexes with self-healing, jelly-like electronic nerves that sense pressure in multiple intensities (not just on/off). ARPA‑H is funding bioprinted “bridge livers” designed to support patients for 22–44 weeks, potentially reducing the need for full transplants. On the compute side, a new multi-physics architecture boosts Fourier-transform speed to ~500500 billion ops/sec, while MIT’s MechStyle lets you restyle 3D models without weakening them. Quantum scaling also took a leap: metasurfaces generating up to 360,000360,000 optical tweezers could help neutral-atom systems push toward 100,000+100,000+ qubits.

🔊 Industry Insights & Updates

Meta is locking in up to 6.66.6 GW of nuclear power by 2035 to fuel AI superclusters—one of the clearest signs energy is now core AI strategy. Robots are also learning dexterous motion from minimal data, and “transparent wood” smart windows can switch from ~28%28% to ~78%78% light transmission as temperatures rise—no electricity required. Cheap 3D-printed optics are improving too, with custom super-resolution lens elements reportedly produced for under 11 dollar.

🧬 BioTech

A “womb-on-a-chip” now mimics early embryo implantation phases, offering a new window into the earliest days of pregnancy. Rice’s CLASSIC approach pairs huge DNA-circuit libraries with ML so scientists can predict new genetic designs, not just test them. MIT also built ingestible pills that signal from the stomach within ~1010 minutes to confirm dosing—promising for transplant meds and long-term therapies.

💡 Products/Tools of the Week

Note67 records meetings with on-device Whisper transcription and local AI summaries via Ollama for privacy-first notes. AcaDraw turns prompts into publication-ready scientific diagrams and editable SVGs in seconds. Feynn generates source-verified competitive and strategic research with scenario simulation for decision support. Diffray speeds up GitHub code reviews with multi-agent analysis focused on real issues (security, performance, concurrency) while avoiding style nitpicks.

🎥 Video Section

Robots are getting more expressive and more real-world-ready—lip-sync demos, humanoid platforms, and agentic operating systems for physical robots are accelerating fast.

The exciting part: we’re watching the shift from “cool prototypes” to systems that scale, self-correct, and integrate into real infrastructure. Biology and computation are increasingly borrowing each other’s tricks—healing, sensing, adapting. If this pace holds, the next few years won’t just add smarter tools; they’ll change what we consider possible to build and deploy. Stay hungry, stay futurish!

🤯 Mind-Blowing

• An innovative electronic 'nerve' system that enables robots to exhibit human-like reflexes has been developed, featuring the ability to detect pressure at varying intensities rather than simple binary responses. Researchers from Northeast Normal University in China created these jelly-based electronic nerves that can autonomously repair themselves after sustaining damage, with sensitivity that increases following an 'injury' and diminishes during 'recovery'. The technology uses memristors with 16 stable levels to encode different pain intensities, ranging from no pain to severe, functioning like a dimmer switch rather than a conventional on/off mechanism. Central to this advancement is gelatin, a collagen-derived protein that conducts ions efficiently and exhibits self-healing properties when heated to 60°C, creating systematic conductive pathways corresponding to the 16 pain levels.

• A $28.5 million grant has been obtained from the Advanced Research Projects Agency for Health (ARPA-H) in the United States to develop functional 3D bioprinted livers intended for transplantation. The initiative, named Immunocompatible Volumetric Engineering (LIVE), aims to tackle the critical shortage of donor organs by producing bioengineered tissue that can help treat acute liver failure. Carnegie Mellon University leads the project with a team comprising specialists in liver stem cells, biomanufacturing, transplantation, and regenerative medicine, collaborating with experts from the University of Pittsburgh, Mayo Clinic, and University of California. The primary objective is to develop a temporary liver that can provide support for approximately two to four weeks while the patient's own liver recuperates, thus potentially eliminating the necessity for a complete transplant. The LIVE team employs a proprietary FRESH 3D bioprinting method that allows them to print soft materials such as collagen and human stem cells into intricate structures using hypoimmune cells engineered to be "universal donor" cells that are undetectable by the human immune system.

• An innovative computing architecture has been developed that enhances processing capabilities by nearly fourfold, paving the way for advancements in fields like artificial intelligence, sensing, brain-inspired computing and communication technologies. Researchers from Peking University ingeniously combined two unique devices to synchronize frequency conversion within a multi-physics-domain framework, creating a flexible system capable of executing intricate tasks including the Fourier transform. The architecture allows multiple computing paradigms to function within their optimal physical environments. the unified system leverages the advantages of two devices involved in frequency generation, modulation, and in-memory computing, ensuring precision while reducing power consumption and increasing the processing speed of the Fourier Transform from approximately 130 billion operations per second to around 500 billion.

• Customizable 3D-printed items that maintain their structural integrity have become possible through an innovative generative AI platform named MechStyle developed by researchers at MIT. The system addresses a prevalent challenge in AI-assisted design where generative models can easily alter objects for aesthetic purposes, but these modifications frequently lead to issues such as snapping, bending, or collapsing post-manufacturing, with most existing AI technologies prioritizing appearance over functionality. MechStyle provides users the option to upload their own 3D models or select from pre-existing templates such as vases and wall hooks, then apply various styles through text or image prompts, with the generative AI model adjusting the geometry while a simulation assesses the impact of these alterations on strength.

• An innovative method has been introduced that could enable quantum computers to expand beyond 100,000 qubits by integrating optical tweezers and metasurfaces. Researchers led by Dr. Sebastian Will and Dr. Nanfang Yu at Columbia University successfully captured 1,000 strontium atoms using metasurfaces—ultra-thin optical devices crafted from millions of nanoscale pixels that manipulate light into thousands of tightly focused laser spots at once. The team produced a 35-millimeter-wide metasurface containing over 100 million pixels capable of generating 360,000 optical tweezers—two orders of magnitude greater than current technologies. Constructed from silicon nitride and titanium dioxide, these metasurfaces can endure extremely intense lasers with optical intensities exceeding 2,000 watts per square millimeter, about a million times more intense than sunlight.
  

🔊 Industry Insights & Updates

• Three major energy partnerships have been announced to deliver up to 6.6 gigawatts of nuclear power by the year 2035, positioning Meta among the largest corporate nuclear energy buyers in American history. The technology powerhouse secured contracts with Oklo, TerraPower, and Constellation to provide clean, dependable electricity for Meta's next-generation AI infrastructure, particularly the Prometheus supercluster situated in New Albany, Ohio. Meta's partnership with TerraPower will support development of two new Natrium units generating up to 690 MW of power as soon as 2032, plus rights to energy from six additional units for a total of 2.1 GW of baseload capacity and 1.2 GW of integrated storage. The company is backing Oklo's establishment of a nuclear technology campus in Pike County, Ohio, expected to become operational by 2030 and contribute 1.2 GW of clean power to the PJM market using Powerhouse reactors that can process both fresh and recycled fuel.

• A novel adaptive motion reproduction system has been created that enables robots to mimic human-like movements with remarkably minimal training data. Researchers in Japan, led by Akira Takakura from Keio University, devised the system using Gaussian process regression to tackle the challenge of robots struggling when object characteristics such as weight, stiffness, or texture are altered. The innovative technique transcends linear models by effectively mapping intricate nonlinear relationships even with a limited dataset, recording human grasping movements on various objects with different stiffness to discern the connection between object characteristics and the forces and positions applied by humans.

• An innovative smart window made from wood has been created that adapts to light conditions, blocks harmful ultraviolet rays, and enhances insulation, all without the need for electricity. Researchers developed the transparent wood by merging modified balsa wood with polymer-dispersed liquid crystals to control visibility and thermal transfer within structures. Laboratory experiments revealed that the material transitions from opaque to clear as temperatures increase, permitting around 28% of visible light to pass through at ambient room temperature while at 40 degrees Celsius this transmittance rises to roughly 78% at a wavelength of 550 nanometers. The transparent wood provides excellent protection against ultraviolet radiation, effectively blocking nearly 100% of UVA rays due to molecular arrangement and π-π aggregation, safeguarding human skin and interior furnishings without significantly diminishing visible light.

• Sophisticated optical components capable of super-resolution imaging have been created using affordable 3D printers and low-cost materials, with each custom lens produced for under $1. Researchers demonstrated that inexpensive, customizable optical elements can achieve imaging that surpasses the conventional diffraction limit by integrating 3D printing, silicone, and UV-curable resin to fabricate multi-element lenses. The team, led by Ralf Bauer from the University of Strathclyde in the UK, constructed a multifocal illumination microscope that resolved biological structures at the nanoscale.

Subscribe now

🧬 BioTech

• A miniature womb on a chip has been developed that accurately simulates how human embryos attach to and penetrate the uterine lining during the crucial first days of pregnancy. Researchers from the Chinese Academy of Sciences created the breakthrough technology by embedding human endometrial cells within gel-like layers inside a microfluidic chip, allowing the cells to organize into a structure that closely resembles actual uterine tissue. The team tested the artificial uterine lining using both real human blastocysts and lab-generated blastoids, which successfully completed all three phases of implantation: initial contact with the uterine surface, stable attachment, and active tissue penetration.

• An innovative method has been introduced that has the potential to transform the way DNA is engineered for medical and biotechnological uses by employing machine learning alongside extensive libraries of DNA designs. Researchers from Rice University unveiled a breakthrough termed CLASSIC—short for "combining long and short range sequencing to investigate complexity"—that enables scientists to generate hundreds of thousands to millions of DNA designs simultaneously, far exceeding previous capabilities. This was the first instance where AI and machine learning could be applied to analyze circuits and accurately predict untested ones because until now no one had been able to create libraries as expansive as theirs.

• Innovative pills capable of transmitting signals from within the stomach have been created by scientists from the Massachusetts Institute of Technology (MIT), allowing the pill to notify when it has been ingested. The technology features a biodegradable radio frequency antenna made of zinc encased within a cellulose particle, with the antenna coiled and housed within a capsule alongside the medication. Once ingested, the capsule’s coating constructed from gelatin and coated with glucose and molybdenum tungsten disintegrates, allowing the drug and RF antenna to be released, with the antenna detecting an RF signal from an external receiver and transmitting a confirmation signal within ten minutes. The innovative system holds promise for tracking transplant recipients who must adhere to immunosuppressive medication regimens, as well as individuals suffering from chronic infections like HIV or tuberculosis who require long-term treatment

  ---

💡Products/tools of the week

• Note67 captures meeting audio and screen content while performing Whisper-powered on-device transcription with advanced speaker separation and echo handling. The application generates private AI summaries using local LLMs via Ollama, ensuring all transcripts and summaries remain exclusively on the user's Windows or macOS machine. Note67 runs completely locally, making it ideal for users who want real-time AI-powered notes with complete data privacy without any cloud dependencies.

• AcaDraw transforms plain-language prompts in English or Chinese into publication-ready scientific illustrations, diagrams, and editable SVGs within seconds. The AI-powered science drawing tool leverages advanced image models including Nano Banana and Seedream 4.0 to deliver high-fidelity strokes and vector-style outputs. Researchers, educators, and students can utilize instant live previews and an interactive chat for iterative refinements, generating precise presentation- and journal-quality figures without requiring manual drawing skills.

• An AI-driven strategic intelligence platform delivers multi-agent, source-verified research on companies, competitors, suppliers, and industries, generating concise briefs while running scenario simulations to recommend traceable, action-ready decisions. Specialized agents map entities, surface real-time signals, and model trade-offs through a decision engine with guided game-theory simulations. Feynn minimizes hallucinations by linking every insight to verified sources, enabling leaders and strategy teams to move quickly with defensible foresight.

• Speeds code reviews and reduces false positives by analyzing GitHub pull requests with full codebase context, diffray's coordinated multi-agent system of specialized AI agents surfaces high-confidence, actionable issues in security, concurrency, data consistency, testing, and performance while deduplicating results and avoiding style nitpicks. The AI-powered code review tool integrates through a GitHub app, enforces "culture as code" team rules, and ensures privacy by not storing code or using it for AI training, helping development teams catch real bugs faster.

  ---

• Columbia Engineering: A Robot Learns to Lip Sync

• Zoomlion Global: ZOOMLION Smart City Vlog! The Real Work-life Balance🚀

• Matrix Robotics: Introducing MATRIX-3

• PNDbotics: Adam-U Ultra Launch: Smarter VLA, Seamless Out-of-the-Box Experience

• LimX Dynamics: Meet LimX COSA｜The First Physical-world-native Agentic OS for Humanoid Robots

• Realbotix: Watch Two Robots Hold a Real Conversation Using Embedded AI | CES 2026

Welcome to this week's edition of Rushing Robotics! 2026—the year AI stepped out of our screens and into our eyelids, brains, and bones. While you were celebrating the new year, labs and factories were quietly birthing technologies that rewrite the rules of what's possible. This issue captures that momentum: from mind-controlled gaming to robots that learn on the job, from quantum chips that solve the unsolvable to nano-magnets that heal while they fight cancer. Let's dive in.

🤯 Mind-Blowing

This week's breakthroughs feel plucked from a sci-fi script, yet they're already performing in real labs and on real patients. We're witnessing the convergence of biological energy harvesting, direct neural interfaces, and autonomous machines that don't just follow orders—they understand context. These innovations don't just push boundaries; they erase them entirely.

🔊 Industry Insights & Updates

The future isn't just being invented—it's being manufactured at scale. This section tracks the momentous shift from prototype to production line, where brain implants become mainstream medical devices, laptops shed their fans for plasma cooling, and industrial giants like Siemens and NVIDIA weave AI into the fabric of global manufacturing. These are the moves that turn breakthroughs into businesses.

🧬 BioTech

We're learning to speak the language of cells themselves. From lungs that breathe on chips to circuits that tame the chaotic noise of cancer, this week's biotech stories reveal a new precision in how we model, understand, and heal the human body. The tools are getting smarter, the interventions more targeted, and the path from lab bench to bedside shorter than ever.

💡 Products/Tools of the Week

Automation is quietly reshaping how we work. Lensmor tracks competitor moves automatically across websites and social media, turning market noise into actionable intelligence. Logic Sheet eliminates repetitive spreadsheet tasks by connecting Google Sheets to your entire tech stack with AI-powered workflows. Devgraph maps your entire codebase into a live knowledge graph so AI agents can truly understand your systems. And Arahi AI launches autonomous teammates that handle multi-step tasks across 2,800+ apps without writing a single line of code.

🎥 Video Section

Seeing is believing. This week's video lineup showcases the robots that will share our workplaces and homes—Atlas demonstrating factory agility, Unitree's daily training routines, and MenteeBot following natural language commands without remote control. Watch these machines move from concept to capability.

The pace of change can feel dizzying, but it's also deeply hopeful. Each breakthrough here represents not just a technical achievement, but a new way to restore mobility, extend health, unlock creativity, and lighten human labor. We're not just building smarter machines—we're building a more capable, more caring world. Stay hungry, stay futurish!

🤯 Mind-Blowing

• A new eye‑tracking system that harvests power from the wearer’s eyelid motion was introduced at Qingdao University, enabling paralyzed patients to control devices through blinks and gaze. Developed by a team of biomedical engineers and neuroscientists, the device attains a 99 % accuracy rate, allowing users to type on a computer screen or steer a wheelchair without moving a single limb. By converting the micro‑voltage generated during an eyelid closure into usable energy, the system functions without external batteries, offering a lightweight and discreet alternative to conventional assistive technologies. In pilot studies, participants successfully navigated a graphical user interface and communicated via an integrated speech synthesis module, showcasing the technology’s practical viability. This innovation represents a major advance in assistive technology, merging bio‑electrical harvesting with precise computer vision to deliver an autonomous and intuitive user experience.

• A brain‑sensing gaming headset debuted at CES, capable of interpreting neural signals while players engage in first‑person shooters. The headset was developed by Neurable alongside HyperX, embedding non‑invasive EEG sensors into a familiar gaming design to deliver live focus and cognitive performance data. In early trials, casual gamers achieved a 43‑ms reaction‑time improvement and a 0.53‑percent accuracy gain, while collegiate esports athletes saw a 3‑percent accuracy rise and a 38‑ms faster response. HyperX’s gaming solutions VP described the product as a tool for players to fine‑tune mental states, and Neurable’s CEO highlighted the shift toward normalizing brain awareness in everyday play. This announcement comes at a pivotal moment for BCI and brain-tracking technologies as they move from research environments into consumer products.

• A new plan to deploy 30,000 humanoid robots each year by 2028 was unveiled, and Hyundai Motor Group said it will use Boston Dynamics’ Atlas robots to automate repetitive factory tasks. The announcement came during CES 2026, where Hyundai also introduced the production‑ready Atlas model, and the company’s Vice Chair, Jaehoon Chang, emphasized that humans will still supervise and train the machines. Hyundai’s strategy, dubbed “human‑centered automation,” aims to reduce physical strain on workers and accelerate production by integrating robots that can lift, assemble, and learn on the fly.

• Google’s latest quantum milestone—Willow, a 105‑qubit chip that finished a random‑circuit benchmark five minutes after launch—breaks the barrier to previously “unsolvable” problems. With repeated error‑correction, the chip can accurately model the quantum mechanics of complex molecules, enabling new pharmaceuticals; predict the behavior of advanced materials for next‑generation electronics; design highly efficient, low‑carbon energy systems; and produce climate forecasts with fine spatial and temporal resolution. Each of these domains had been constrained by classical computing limits; Willow lifts those constraints, giving Google a decisive advantage in tackling global challenges.

• Nano‑magnets have been found capable of defeating bone cancer and aiding healing, as announced by researchers from Brazil and Portugal. Their team, led by Dr. Angela Andrade of the Universidade Federal de Ouro Preto, synthesized iron‑oxide nanoparticles coated with bioactive glass that, when exposed to an alternating magnetic field, burn cancer cells from the inside while simultaneously encouraging new bone growth. This dual‑action approach not only eradicates tumors but also restores the damaged bone, a significant advance over existing treatments that only target the disease. By combining high magnetization with strong bioactivity in a single material, the researchers hope to move this promising technology toward clinical trials and eventually offer patients a minimally invasive, two‑in‑one solution for bone cancer.

🔊 Industry Insights & Updates

• Neuralink announced that it will launch high‑volume production of its brain‑computer interface devices in 2026, moving to an almost entirely automated surgical procedure, a move Elon Musk disclosed on X after months of regulatory review. The implant, designed to help people with spinal cord injuries and other severe neurological conditions, already enabled a first patient to play video games, browse the internet, and control a laptop cursor using thought alone. Musk’s statement highlighted the company’s plan to expand its clinical trials to dozens of patients worldwide, while also securing a $650 million funding round in June that will support mass manufacturing.

• YPlasma revealed the world’s first laptop cooled by DBD plasma actuators, a technology that replaces traditional fans with a thin film of cold plasma generating high‑velocity ionic wind; CEO David García Pérez highlighted the system’s noiseless performance and safety advantages, noting that the company’s engineering teams in Madrid and Newark engineered a 200‑µm‑thick actuator that can be integrated directly onto heat sinks, chassis walls, or internal components for ultra‑thin, high‑performance devices.

• At CES 2026, Siemens and NVIDIA announced a new Industrial AI Operating System that will integrate artificial intelligence across design, engineering, manufacturing, operations, and supply chains. The partnership, led by Siemens President Roland Busch and NVIDIA CEO Jensen Huang, will enable factories to simulate changes virtually, test improvements in real time, and deploy validated insights directly onto production lines. A fully AI‑driven manufacturing site is slated to launch in 2026 at Siemens’s Erlangen electronics factory, while the duo plans to roll out GPU‑accelerated simulation tools, generative physics models, and AI‑powered copilots for workers, scaling the platform to customers such as Foxconn, Hyundai, KION, and PepsiCo.

• LG turned a spotlight on its “AI in Action” roadmap at CES 2026, presenting a lineup that turns digital intelligence into physical, helpful solutions. The company’s CEO, Lyu Jae‑cheol, launched the event by asking the industry to imagine AI stepping out of the screen and into real life. He then explained that LG’s strategy relies on three core pillars: superior device technology, a seamless ecosystem, and the expansion of AI solutions beyond the home. It also highlighted the CLOiD humanoid robot, built to learn and adapt to a home environment, capable of folding laundry and adjusting heating and cooling, with a focus on safety and stability. The message wrapped up by envisioning a future where AI quietly improves everyday life, making it more meaningful and human.

Subscribe now

🧬 BioTech

• A breathing lung‑on‑chip was engineered from a single donor’s cells, allowing researchers to observe tuberculosis (TB) infection in a realistic, human‑like environment. The new model, announced on January 1, was created by scientists at the Francis Crick Institute and the Swiss company AlveoliX, who used induced pluripotent stem cells to build a lung tissue that contracts and expands with simulated breathing. By integrating the donor’s own immune cells (macrophages) and introducing Mycobacterium tuberculosis, the team could track the pathogen’s early progression and the host’s cellular response in real time. This breakthrough, promises to refine personalized treatment plans and reduce reliance on animal models for respiratory disease research.

• Researchers from KAIST and POSTECH have unveiled a mathematical device that can tame random fluctuations inside single cells, a development that could halt cancer’s ability to relapse after chemotherapy. The Noise Controller, a feedback circuit that senses the very “noise” in protein levels rather than just their averages, was designed by Professor KIM Jae‑Kyoung. By employing protein dimerization and targeted degradation, the team proved in computer simulations that the controller can reduce failure rates in E. coli DNA‑repair pathways from 20 % to 7 %, demonstrating that outlier cells that resist drugs can be coaxed into normal behavior. The study, signals a shift from population‑wide regulation to single‑cell precision control, offering a promising strategy for smart microbes, synthetic biology, and especially for overcoming drug resistance in cancer therapy.

• Spark Biomedical unveiled the OhmBody neurostimulation wearable, a non‑invasive, hormone‑free device that targets nerves around the ear to curb heavy menstrual bleeding. Katherine Reil, the company’s spokesperson, explained how the device uses transcutaneous auricular neurostimulation of the vagus and trigeminal nerves to activate platelets in the spleen, reducing blood loss by more than 50 % and shortening periods by roughly 20 % in a 16‑participant trial. While still seeking FDA clearance, Spark Biomedical plans to scale the product through e‑commerce and sports‑wear partnerships, positioning OhmBody as a pioneering wellness‑to‑medicine solution for women’s health.

  ---

💡Products/tools of the week

• Competitor moves get tracked automatically with Lensmor, an AI-driven competitive intelligence platform that monitors competitor websites, Reddit, and LinkedIn 24/7. Lensmor uses machine learning to spot website changes, A/B tests, job moves, and sentiment shifts, then prioritizes the highest-impact signals and turns them into concise, role-specific action plans. Product, growth, sales, and investor teams use Lensmor to replace manual monitoring, cut noise, convert competitor activity into clear next steps, and stay ahead of market shifts.

• Repetitive spreadsheet work gets automated with Logic Sheet, a Google Sheets add-on that runs workflows through triggers, conditions, and actions like emails, Slack messages, webhooks, sheet updates, and API integrations. Logic Sheet connects Google Sheets to tools including Airtable, Notion, HubSpot, and Mailchimp so data and tasks move automatically between systems. Non-developers also use Logic Sheet’s built-in AI agents powered by OpenAI/Gemini to follow plain-English instructions for multi-step tasks such as scraping, summarizing, generating content, enriching data, and drafting personalized outreach—eliminating manual work, reducing errors, and saving hours without coding.

• Devgraph is an AI-powered ontology engine that automatically discovers and maps relationships across your code, infrastructure, tickets, and chats into a live knowledge graph so LLMs and AI agents can understand and act on your real systems in real time; use it to run natural-language queries, perform dependency/impact analysis, surface tribal knowledge, speed onboarding, and automate workflows (create tickets, trigger deploys) while retaining control by bringing any LLM—cloud, self-hosted, or air-gapped—for privacy, cost, and performance.

• Autonomous workflows get launched from simple prompts with Arahi AI, an agentic, no-code automation platform that runs AI teammates powered by Agent NEO. Arahi AI connects to 2,800+ apps and operates 24/7, creating and configuring automations that can trigger on events like incoming emails or support tickets. Using large-memory context, template-driven logic, and models tuned for near-zero hallucinations, Arahi AI executes multi-step tasks reliably and surfaces real-time insights so teams can replace repetitive work across customer service, sales, marketing, finance, and HR with predictable, lower-cost automation and faster time-to-value than traditional tools.

  ---

• Cybernews: Atlas Robot First Look - Boston Dynamics at CES 2026

• Boston Dynamics: Atlas POC Video | Boston Dynamics

• Boston Dynamics: Atlas | Product Features | Boston Dynamics

• Unitree Robotics: Unitree Humanoid Robot Daily Training 🥳

• UBTECH Robotics: UBTECH Walker S2 Tennis Rally: A Perfect Stroke for 2026

• Menteebot by Mentee Robotics: MenteeBot follows verbal instruction (No Teleoperation)

• AGIBOT: AGIBOT Launches SOP: Scaling General-Purpose Robots in the Real World

Welcome to this week's edition of Rushing Robotics! Merry Christmas to you and yours—hope you’re getting a little rest, a little joy, and maybe even a little time to geek out about what’s coming next. While many are unwrapping presents this week, the future is unwrapping itself at breakneck speed. From brain-inspired chips that compost when you're done with them to robots attending school alongside human teachers, this week's breakthroughs blur the line between science fiction and Friday morning headlines.

🤯 Mind-Blowing

Researchers built an artificial synapse from crab shells, bean fibers, and plant stems that learns like a brain then safely decomposes—directly tackling the e-waste crisis. Chinese scientists developed light-based AI chips that beat NVIDIA GPUs by over 100x on specialized tasks, computing with photons instead of electrons. In Beijing, a two-story robot "school" opened where humanoid androids major in factory work, home chores, and elderly care, learning through relentless repetition like children. And in a delightfully macabre twist, scientists transformed mosquito mouthparts into 3D-printing nozzles capable of lines thinner than a white blood cell.

🔊 Industry Insights & Updates

The business world is racing to turn these lab dreams into market reality. STMicroelectronics signed the semiconductor industry's first humanoid robot supply agreement, bringing Oversonic's RoBee workers directly into chip fabs worldwide. NVIDIA made a $20 billion strategic move, licensing Groq's inference chip IP and acquihiring key staff without buying the company outright—cementing its dominance while absorbing fresh talent. Surgeons in Kuwait and Brazil performed the world's longest remote robotic surgery link at 7,456 miles apart, with just 199 milliseconds of latency. Meanwhile, a new lensless imaging technique called MASI achieves sub-micron resolution without optics, turning sensor arrays into virtual giant apertures through software alone.

🧬 BioTech

Biology continues to reveal its deepest secrets while offering powerful new tools. The 4D Nucleome Project mapped over 140,000 DNA looping interactions, creating the most detailed 3D models yet of how chromosome folding controls gene expression. Researchers engineered bacterial patterns that hide messages until specific biochemical triggers reveal them, creating living encryption for anti-counterfeiting applications. And a robotic "click chemistry" platform synthesized and screened 700 metal-based antibiotic candidates in a week, discovering an iridium-based compound that kills MRSA-like bacteria while remaining non-toxic to human cells—potentially bypassing traditional resistance mechanisms.

💡 Products/Tools of the Week

Practical AI tools are making creative and business workflows faster than ever. Patternaly transforms text prompts into seamless, print-ready patterns in 19 art styles, generating 1–4 variations in 10–30 seconds with free and commercial licensing options. Spell from Spline creates full 3D worlds from text or images in minutes, producing editable Gaussian splatting/NeRF scenes ready for web, VR/AR, and games. Guideless automatically converts browser click-throughs into polished, branded video guides without manual recording or editing—perfect for product and support teams. We.Inc unifies websites, campaigns, and sales workflows in a single no-code dashboard, combining AI chatbots, social scheduling, and creative tools to replace multiple subscriptions.

🎥 Video Section

Enough reading—let's see these concepts in motion. This week's videos show humanoid robots wrapping gifts, learning to fold towels, cooking like humans, and Figure 03 showing off some wild new capabilities.

These threads—biodegradable computing, photonic AI, robot apprentices, and living cryptography—are weaving together into a tapestry of possibility that feels fundamentally different from incremental progress. We're watching the convergence of sustainability, intelligence, and biology into something that doesn't just change what we build, but how we exist alongside our creations. Stay hungry, stay futurish!

P.S. We're taking a short holiday break—your next dose of the future arrives in two weeks.

🤯 Mind-Blowing

• A plant fiber-based artificial synapse was built to function like a tiny brain-inspired chip while safely decomposing after use, directly targeting e-waste and sustainability challenges. The UNIST research team in South Korea used biodegradable polymers and bio-derived ingredients such as shells, beans, cellulose acetate from plant stems, and chitosan from crab to form a small, layered device. When stimulated, the synapse releases sodium ions that behave like neurotransmitters, binding inside the structure so signals can persist without continuous power and enabling synaptic plasticity behaviors linked to learning. In testing described in the report, the device maintained memory for close to 6,000 seconds and was used to drive a basic heat-triggered robotic reflex, then was positioned as a route toward eco-friendly electronics that can return to nature.

• Light-based AI microchips were reported to beat NVIDIA GPUs by over 100x on certain specialized generative tasks, signaling a fast, efficient alternative architecture rather than a drop-in replacement. Chinese researchers developed photonic chips such as ACCEL and LightGen that compute with photons through optical interference, unlike NVIDIA GPUs (including A100) that rely on electrons moving through transistors and deliver flexible, program-driven computing at high power and heat cost. The report says ACCEL is a hybrid photonic-analog electronic design linked to Tsinghua that can be made using established fabrication processes, including older SMIC technology, and has demonstrated performance on predetermined analog math suited to tasks like image recognition and low-light vision. LightGen, described as a fully optical chip reportedly includes more than 2 million photonic “neurons” and targets image-generation-style workloads such as style transfer, denoising, and 3D manipulation with far lower power use.

• A robot “school” was launched in Beijing to train humanoid robots for factory work, home chores, and public-service roles using real-world practice environments. China opened the Phase II Beijing Humanoid Robot Data Training Center in Shijingshan district, a two-floor facility that replicates production lines and living spaces where robots practice tasks like sorting coils, picking parcels, cooking, and organizing bedrooms. The center’s director Zhu Kai said each humanoid is paired with two human trainers who repeatedly demonstrate tasks, arguing that robots—like children—need extensive repetition across scenarios to build functional intelligence. The project targets a key bottleneck in humanoid robotics: scarce standardized real-world data, and the facility is expected to generate millions of high-quality data entries annually while linking to similar centers in multiple cities. Robots can pick majors across manufacturing, home applications, elderly care, and 5G-integrated scenarios, and the report says trained systems have learned 20+ skills with success rates above 95%.

• A “necroprinting” method was introduced that turns mosquito feeding tubes into ultra-fine 3D-printing nozzles, enabling line widths down to about 20 microns—slightly narrower than a white blood cell. A collaborative team from McGill University and Drexel University repurposed the proboscis of female mosquitoes as a biodegradable, high-resolution outlet that reduces clogging and pressure buildup thanks to its naturally evolved fluid-transport design. The researchers extracted proboscides from euthanized mosquitoes from ethically managed Drexel lab colonies, then detached each tube under a microscope and mounted it onto a standard plastic dispenser tip using resin. The team said the approach could reduce cost and waste compared with specialized metal or glass micro-nozzles and may support biomedical microfabrication and even delicate semiconductor handling.

• Over 140,000 DNA looping interactions were charted to build one of the most detailed 3D maps yet of how human chromosomes fold inside the nucleus, offering new insight into gene control and replication. A team in the 4D Nucleome Project mapped chromosome architecture in two cell types—human embryonic stem cells and fibroblasts—because genome sequence alone does not explain how genes are switched on or off. Using multiple genomic assays to measure which DNA regions come into contact, the researchers integrated the results into an Integrative Genome Modeling (IGM) platform that generated 1,000 distinct single-cell 3D genome models. The dataset cataloged more than 141,365 regulatory looping interactions in stem cells and 146,140 in fibroblasts, showing how genes can interact with distant regulatory elements in three-dimensional space. The team also trained deep learning models to predict genome folding directly from DNA sequence, aiming to forecast how disease-linked variants may reshape 3D structure and affect function.

🔊 Industry Insights & Updates

• Humanoid robots were set to enter chip factories after STMicroelectronics partnered with Oversonic Robotics to roll out RoBee for production and logistics operations. Announced on December 22, the agreement is framed as a first for the semiconductor sector: integrating humanoid systems directly into fab workflows across multiple ST sites worldwide. Oversonic executives described the move as a milestone that validates their vision for industrial and healthcare markets and demonstrates confidence in meeting semiconductor-grade standards for precision, safety, and uninterrupted operations. RoBee is presented as a certified humanoid for both industrial and healthcare contexts, built to work autonomously in complex environments using AI perception, real-time decisions, and multimodal interaction, including pick-and-place and factory data-gathering tasks that feed into production planning and monitoring.

• Groq’s assets and IP were brought into a $20 billion non-exclusive licensing agreement, paired with an acquihire of key staff, expanding NVIDIA’s inference push without buying Groq outright. The report says NVIDIA will license Groq’s intellectual property and hire several top Groq employees. Groq is portrayed as a rival AI chip maker focused on LPUs—ASIC-style processors optimized for inference—arguing they can be more efficient than general-purpose GPUs for certain workloads. Jensen Huang was quoted saying NVIDIA will integrate Groq’s low-latency processors into its AI factory architecture to support more inference and real-time tasks, and emphasized that the agreement is not an acquisition of Groq as a company. Under the new structure described, Jonathan Ross and Sunny Madra move to NVIDIA, Simon Edwards becomes Groq CEO, and GroqCloud keeps running as Groq remains independent.

• A world-record robotic surgery link was completed across 7,456 miles when operating rooms in Kuwait and Brazil performed live procedures with surgeons controlling instruments almost in real time. On September 23, 2025, the connection spanned 12,034.92 kilometers between Jaber Alahmad Hospital in Kuwait and Hospital Cruz Vermelha in Brazil using robotic platforms over a secure, high-bandwidth international network. The report says the network delivered an average latency of 199 milliseconds, 80 Mbps bandwidth, and 0.19% packet loss, enabling precise remote control without noticeable delays. Two procedures were carried out to prove two-way capability, with TAPP (transabdominal preperitoneal) hernia repairs performed first by a Kuwait-based team on a Brazil patient and then by Brazil-based surgeons on a Kuwait patient.

• A new software-driven optical technique achieved sub-micron imaging without lenses by separating measurement from synchronization, turning sensor arrays into a virtual giant aperture. Developed at the University of Connecticut by Professor Guoan Zheng and colleagues, the system—called Multis Aperture Synthesis Imaging (MASI)—takes cues from the Event Horizon Telescope’s synthetic aperture method, but solves visible-light alignment problems by shifting synchronization into computation. Coded sensors placed at points in a diffraction plane each capture raw diffraction patterns, after which algorithms reconstruct the complex wavefield per sensor, propagate and pad the data, and iteratively phase-synchronize the set for coherent merging. The approach removes the need for strict interferometric alignment that has historically limited optical synthetic aperture systems and enables high resolution, wide field of view imaging from several centimeters away.

Subscribe now

🧬 BioTech

• The most detailed “4D” maps yet of how human DNA folds and moves over time were created, revealing how genome shape helps control which genes switch on or off. Scientists at Northwestern University, working with the international 4D Nucleome effort, built high-resolution views of the genome’s three-dimensional organization across time, focusing on human embryonic stem cells and fibroblasts to capture development- and function-relevant states. The research describes DNA not as a linear string but as loops, compartments, and domains that physically position genes near or far from regulatory elements, influencing cell identity, development, and disease processes. By integrating multiple complementary genome-mapping technologies into a single dataset, the team reported identifying more than 140,000 chromatin loops per cell type and mapping the molecular components that anchor those loops and shape gene regulation. The study also introduced computational tools that can predict genome folding from sequence, aiming to help researchers anticipate how disease-linked variants may reshape 3D organization and affect gene expression.

• Bacterial patterns were created that hide messages until the right biochemical trigger reveals them, offering a new approach to secure information and anti-counterfeiting. Researchers reported a multilevel encoding system in Advanced Functional Materials that uses specially designed photodynamic nanoparticles and light exposure through a photomask to kill bacteria in illuminated regions while leaving shaded regions alive to form precise “living codes.” The nanoparticles combine a light-sensitive molecule (MeO-TSP), fatty acids that intensify reactive oxygen species production under white light, and a protective polymer shell that helps bring the particles close to bacterial membranes for efficient cell destruction. Once patterned, different bacterial species reveal different messages depending on the chromogenic substrate provided, enabling Morse-code and QR-code style outputs, including decoys that only decode correctly with the proper biochemical cue.

• Hundreds of metal-based antibiotic candidates were synthesized and screened in about a week using a robotic “click” chemistry platform, producing a standout iridium-based lead that kills bacteria while staying non-toxic to human cells. A team led by Dr Angelo Frei at the University of York’s Department of Chemistry used automation to combine almost 200 ligands with five different metals, generating over 700 metal complexes far faster than months of manual lab work. After testing all 700 for antibacterial activity and toxicity, the researchers identified six potential lead compounds and highlighted one iridium complex with strong activity against bacteria including MRSA-like strains and a high therapeutic index for further development. The work argues that three-dimensional metal complexes can interact with bacteria differently than conventional “flat” carbon-based antibiotics, potentially bypassing existing resistance mechanisms, and cites CO-ADD data suggesting metal complexes can have a higher antibacterial hit rate without added toxicity.

  ---

💡Products/tools of the week

• Patterns were produced as seamless repeats using Patternaly’s AI Pattern Generator for Seamless Designs, a web-based tool that transforms basic text prompts into print-ready designs. Patternaly lets users pick from 19 art styles, generate 1–4 variations in about 10–30 seconds, preview patterns on product mockups, and export high-resolution PNG files. Patternaly also offers a free option, while paid plans add commercial rights for broader usage.

• Scenes were generated in minutes from text prompts or images with Spell, an AI-powered scene generator from Spline that creates full, stylized 3D worlds with renderable volumetric outputs such as Gaussian splatting/NeRF. Spline built Spell to produce editable, production-ready environments that can be refined and exported for web, VR/AR, games, and prototypes through its browser-based editor.

• Walkthrough videos were created automatically from browser click-throughs with Guideless: AI Documentation & Video Guides, turning captured flows into polished, branded video guides in minutes. Guideless built the product so product, support, and enablement teams can produce consistent how‑to videos without manual recording, editing, or scripting.

• Websites, campaigns, and sales workflows were unified in one dashboard with We.Inc, a 5‑in‑1 no‑code software suite positioned to help businesses build, market, and sell from a single platform. We.Inc combines core capabilities like website creation, AI chatbots for lead capture/qualification, social scheduling, and AI creative tools to reduce the need for multiple subscriptions.

  ---

• Humanoid: What happens if a Humanoid robot wraps your Christmas gift?

• Kinisi Robotics: Ai Robotics - How Do You Teach a Robot to Fold a Towel

• HouseBots: From Player Piano to Master Chef: Memo Learns Like a Human

• XRoboHub: 🎄🤖 Figure 03 Gone Wild!

Welcome to this week's edition of Rushing Robotics! This week’s theme: intelligence is getting smaller (microrobots, stacked chips), broader (robot ecosystems), and more biological (cell-size genetics, tunable proteins). Let’s dive in.

🤯 Mind-Blowing

Microrobots at near-microbe scale are now fully programmable and autonomous, stacked-chip architecture hints at the next big leap in AI hardware, and non-coding RNA just got a serious upgrade as a potential master control for cell size. Also: someone shipped a playable game they claim was built 100% with AI—an early sign of how fast “vibe coding” is compressing production cycles.

🔊 Industry Insights & Updates

China is building robot-friendly public training zones, Unitree is pushing a humanoid “app store” model, and on-device AI keeps accelerating toward data-center-class capability. At the same time, the Google–NextEra push shows the real limiter is still power—AI progress is becoming energy strategy.

🧬 BioTech

Researchers can now tune protein dosage in specific tissues across an animal’s whole life, cell-sheet tissue engineering got dramatically faster, and embryo development is becoming predictable with deep learning—opening doors to earlier detection of subtle developmental failures.

💡 Products/Tools of the Week

Superway speeds up trend discovery and forecasting from digital noise, SociaVault delivers structured social data via API for analytics and ML, Subject.so turns stories into coherent long-form video with continuity, and TextDeck generates polished PowerPoint decks from prompts in about a minute.

🎥 Video Section

Humanoids, safety concerns, biomimetic robots, and a clever gripper design—solid lineup if you want to see where embodied AI is heading.

The future feels closer because these advances stack: better chips enable better robots, better tools accelerate building, and better biology controls expand what’s possible in medicine. The next leap won’t be one headline—it’ll be the compound effect of many. Stay hungry, stay futurish!

🤯 Mind-Blowing

• Microrobots were introduced at roughly the scale of microorganisms, with each unit described as nearly invisible to the human eye and measuring about 0.3 by 0.05 millimeters, placing them in the size range of bacteria and unicellular organisms. The University of Pennsylvania team built the robots and integrated computing hardware from the University of Michigan, creating what they claim are the smallest fully programmable autonomous robots to date. The robots can navigate complex paths, respond to temperature changes, and coordinate as groups, demonstrating combined mobility, sensing, and computing at extreme miniaturization.

• A new discovery suggests cell size is directly controlled by a gene in the non-coding genome, offering a clearer genetic explanation for why cells don’t grow too large or stay too small. A SickKids-led research team reported that the long non-coding RNA CISTR-ACT serves as a primary regulator of cell size, shifting scientific thinking about the biological importance of non-coding DNA. The scientists said CISTR-ACT influences genes tied to cell growth, structure, and adhesion, meaning it may shape how cells develop and organize across multiple tissues rather than acting in a narrow, cell-specific way. In preclinical experiments, lowering CISTR-ACT levels increased cell size—such as enlarged red blood cells and observed brain-structure changes—while boosting CISTR-ACT produced smaller cells, reinforcing its direct role in size control. The team further explained that CISTR-ACT helps the protein FOSL2 bind to and regulate growth-related genes, especially during brain and bone marrow development, and they noted potential relevance to diseases where cell size matters, including cancer and anemia.

• A major stacked-chip advance was presented by a US-led research team that says it signals a new chapter in chip manufacturing and could help set the direction for next-generation AI hardware. Stanford professor Subhasish Mitra, the project’s principal investigator, said breakthroughs like this are important on the path toward the far larger performance jumps—on the order of 1,000-fold—that future AI systems may demand. The team reports that the chip’s architecture achieves about a tenfold improvement compared with conventional 2D chips by combining an unprecedented density of vertical connections with a tightly integrated mix of memory and processing units. Senior author Tathagata Srimani, now at Carnegie Mellon University, compared the idea to elevator banks in a tall building: many high-speed “elevators” can move large numbers of “riders” (data) between floors (layers) at the same time, easing the traffic jams that hold back flat designs.

• Proton mobility in metal oxides got a boost with a new MIT physical model that better predicts how protons move through many oxide materials, a step that could help develop room-temperature proton-conducting technologies as alternatives to lithium-based systems. Researchers led by MIT professor Bilge Yildiz say the model highlights why today’s proton conduction in these materials typically requires temperatures above 752°F (400°C), and points to pathways for making inorganic, scalable proton conductors that work at lower temperatures.

• A fully playable game was unveiled that its creator claims was made entirely with AI. Developer Grolaf, also known as Crunch Fest, introduced Codex Mortis, describing it as the “first-ever” game created through 100% AI-driven development and “vibe coding,” with assets, programming, and background music generated by AI. Inspired by Vampire Survivors, the title is positioned as a necromantic survival bullet hell where players fight demonic hordes, collect ancient pages to progress, and unlock abilities. The project was reportedly built in three months, with ChatGPT playing a major role in shaping the artwork and helping maintain visual consistency, while animations leaned on custom shader techniques produced with Claude Code.

🔊 Industry Insights & Updates

• A province-wide training network for embodied intelligent robots is being organized in Guangdong, with Shenzhen planning the country’s first city area built to be explicitly robot-friendly. The concept, branded the “1+1+N” Guangdong Embodied Intelligence Training Ground, was introduced by officials as Shenzhen unveiled the Shenzhen Embodied Intelligence Demonstration Zone during the Greater Bay AI and Robotics Industry Conference ending December 14. Guangdong’s plan centers on a main training site and management hub, then extends into the Shenzhen demonstration zone where robots can practice on public streets in everyday situations, and finally expands through multiple sector- and city-specific “N” sub-training facilities designed to collaborate with local governments and industries.

• Humanoid robots got a new “app store” experience as Unitree Robotics unveiled what it says is the world’s first app store dedicated to humanoid robots, aiming to make robots as customizable as smartphones. Unitree, based in Hangzhou, presented the platform through videos shared on YouTube, X, and RedNote, showing how users can access and upgrade robot functions directly from mobile devices.

• The Tiiny AI Pocket Lab was introduced as a compact device capable of running up to 120 billion-parameter LLMs entirely on-device, positioning Tiiny AI as the company bringing “data-center-level” AI capabilities to everyday users in an offline format. Tiiny AI said the device was officially launched on December 10 and framed it as a response to rising energy costs, sustainability concerns, and privacy threats linked to cloud-based AI.

• A major buildout was unveiled to develop several gigawatt (GW) AI data center campuses across the US as part of an expanded partnership between NextEra Energy and Google Cloud. The companies said the new initiative extends an existing collaboration totaling about 3.2 GW of capacity that is already operating or under contract. They also pointed to recent contracts intended to add 600 megawatts (MW) of renewable energy capacity to Oklahoma’s grid to help meet rising power demand tied to large-scale computing.

Subscribe now

🧬 BioTech

• Protein levels were dialed up or down in specific tissues for an animal’s entire lifespan by a research team addressing a long-standing limitation in biology where scientists could remove proteins or switch off genes but could not finely tune dosage in chosen tissues over time. The researchers demonstrated the technique in the nematode worm C. elegans, adjusting protein amounts in the intestine and neurons while the animals continued to live, feed, and develop normally, which is essential for studying long-running processes like aging without disrupting basic biology. They argue this matters because aging and many diseases depend on continuous communication between organs, and a protein can be beneficial in one tissue yet harmful in another—effects that traditional genetic approaches often cannot separate.

• A rapid tissue-engineering technique was created that produces mechanically strong, peelable cell sheets in roughly five hours using basic lab tools such as a silicone dish, spatula, and tweezers. The work comes from researchers at McMaster University, led by professor and biomedical engineering co-director Pavi Selaganapathy, and it was reported as a major speed-up over scaffold-free methods that can take up to two weeks. The team says the cell sheets may support applications including burn treatment, organ repair, and other regenerative medicine uses.

• Single-cell embryo behavior was made forecastable through an MIT-built deep-learning model that predicts where cells will go and what they will do as the embryo forms. The researchers trained it on high-resolution videos of fruit fly embryos starting with roughly 5,000 cells, and the model learns from each cell’s position, its neighborhood, and whether it is folding or dividing across time. Focusing on gastrulation, which happens over about an hour with minute-by-minute rearrangements, the model reached about 90% accuracy in predicting how cells would fold, move, and reorganize, and it could also predict the exact minute specific events would occur. It’s a meaningful advance because early developmental coordination must be extremely precise, and the article links small early missteps to later serious outcomes, including abnormalities and early-onset diseases such as asthma and cancer; the team also says the model could help identify subtle dynamic signatures in live tissue that improve diagnosis or drug-screening tests.

  ---

💡Products/tools of the week

• Emerging trends and shifting consumer behaviors are being surfaced with unprecedented speed through Superway, a new AI-powered trend research oracle that offers a comprehensive suite of intelligence tools. By employing its sophisticated Oracle AI, the platform analyzes vast amounts of digital noise to score sentiment and forecast the trajectory of potential market movers. The system is built around actionable AI-backed workflows, including an AI copilot that assists users in navigating complex data landscapes. With features such as SuperSense for initial discovery and SuperSeed to expand fleeting thoughts into robust concepts, Superway empowers innovation and marketing teams to spot opportunities earlier than the competition.

• Real-time, structured social data from 25+ networks is extracted and delivered as JSON-ready feeds through SociaVault, a developer-first REST API built to remove the headache of managing scrapers. SociaVault pulls profiles, posts, comments, trends, and ads so teams can quickly build analytics dashboards, monitor competitors, and—critically—power AI/ML pipelines for training sentiment models, recommendation engines, trend predictors, and content classifiers. SociaVault pairs this data access with high concurrency so developers can scale collection without bottlenecks. SociaVault also supports pay-as-you-go credits and includes a free trial to help teams validate use cases fast.

• Ideas are turned into cohesive storyboards and finished videos through Subject.so, an AI-powered film-creation platform built to help users produce long-form content without animation or editing skills. Subject.so uses narrative AI plus proprietary image and video models to generate plots, maintain consistent characters, and preserve scene continuity so stories feel unified from start to finish. Subject.so also produces stylized visuals and enables one-click video synthesis to move from concept to video quickly. Subject.so is positioned for creators, educators, and marketers who want visually consistent, monetizable stories on a faster production cycle.

• Simple text prompts are converted into fully designed PowerPoint decks within about 60 seconds using TextDeck, an AI-powered slide builder that eliminates manual presentation design. TextDeck automatically structures content, applies professional layouts with proper visual hierarchy, and generates charts and data visualizations so slides feel polished and cohesive. TextDeck helps founders, PMs, marketers, consultants, and sales teams save time by producing presentation-ready decks without requiring design skills or lengthy production cycles. TextDeck is particularly useful for teams that need to prototype pitches quickly or generate consistent presentations on demand.

  ---

• XRoboHub: 🔥🤖 NOETIX #Hobbs W1 Unveiled: The All-Rounder with 6-DoF

• InsideAI: ChatGPT in a robot shows we are close to disaster

• Suzumori Endo Robotics Laboratory: Dog Musculoskeletal Robot🐕

• Robotics & AI Lab - CUHK Shenzhen: SnailBot - Enhancing Connection Strength Through Holey Sphere and Gripper Mechanisms

Welcome to this week's edition of Rushing Robotics! This week is all about the moment when breakthroughs stop sounding like demos—and start looking like deployable reality. From bold AGI claims and robots learning on their own, to aviation and wearables inching closer to everyday life, the pace isn’t slowing down.

🤯 Mind-Blowing

A Tokyo startup is making one of the biggest claims in modern computing—an “AGI-capable” model that can autonomously learn new skills safely and efficiently, with robotics as the proving ground. In the same breath, we’re watching the “impossible” become manufacturable: the first true electric flying car entering production, a prosthetic hand that intuitively grasps with AI assistance, humanoid robots running vertical farms, and a sprawling optical network that could compress AI training cycles from months to days.

🔊 Industry Insights & Updates

The competition to own the next interface is heating up again, with Google preparing a return to smart glasses—lighter, AI-infused, and positioned squarely against Meta and Apple. Meanwhile, quantum computing takes a tangible scaling step with a 10,000-qubit architecture push, wearable tech gets a practical upgrade with iron-on conductive circuits, and all-electric aviation moves from concept to field testing on real regional routes.

🧬 BioTech

Brain tech continues to miniaturize and accelerate: a single-chip wireless interface aims to make neural sensing thinner, safer, and far more capable. We also have a clever “resettable” blood test approach that could make brain gene activity readable with less noise, plus a landmark in surgical robotics—robot-assisted cataract procedures demonstrating precision that could reshape one of the world’s most common operations.

💡 Products/Tools of the Week

This week’s tools are all about shipping faster with less friction. From AI standups and prompt-performance insights inside your editor, to full-stack “vibe-coding” platforms that generate real apps you own, the theme is leverage. We’re also seeing AI take on the unglamorous but critical work—like automatically generating end-to-end browser tests for every pull request, and turning simple prompts (or even voice) into deployed, shareable applications.

🎥 Video Section

If you want to see the future rather than just read about it, this week’s videos deliver: a DeepMind robotics lab tour, humanoids navigating debris, Boston Dynamics unpacking Atlas’ movement design, and a six-armed humanoid that looks like it walked out of industrial science fiction.

The most exciting part isn’t any single headline—it’s how these threads are starting to connect. Better models, better sensors, better materials, and better manufacturing are turning “lab miracles” into repeatable systems. As these technologies mature, they’ll move from novelty to infrastructure, quietly changing what work, health, and mobility look like. Stay hungry, stay futurish!

🤯 Mind-Blowing

• A claim of unlocking the world’s first Artificial General Intelligence model has been made by a startup in Tokyo. This bold assertion comes from Integral AI, a company founded by former Google executive Jad Tarifi, which states its system can learn new tasks without prior datasets or human help. The firm defines AGI through three specific criteria: autonomous skill acquisition, safe mastery, and energy efficiency comparable to humans. To prove its technology, the company reportedly conducted trials where robots learned new abilities completely on their own, emulating the human neocortex. While tech giants have long chased this milestone, Tarifi chose Japan for its robotics leadership to develop what he calls a “fundamental leap” beyond current tech. Whether this truly constitutes AGI remains a topic of debate, but the company views it as a new chapter for civilization, aiming ultimately for embodied superintelligence.

• Production of the world’s first true electric flying car has officially begun in California. This historic move transitions the concept of personal air travel from science fiction to a tangible reality after a decade of research. Alef Aeronautics is responsible for this breakthrough and has started the meticulous hand-assembly of the initial units at their Silicon Valley facility. The vehicle, known as the Model A, offers a unique capability to drive on public roads and take off vertically without a runway, unlike other contemporary aircraft. With a driving range of 220 miles and a flight capacity of 110 miles, it addresses the need for sustainable and versatile transport. Jim Dukhovny, the company’s CEO, announced that the team is on schedule to meet the demand represented by 3,500 pre-orders totaling $1 billion. Early units will go to a select group of customers who will provide feedback to help refine the manufacturing process before mass production begins.

• A major step forward in prosthetic technology has occurred with the development of a bionic hand that uses artificial intelligence to grasp objects intuitively. This innovation addresses the cognitive burden often placed on amputees, who typically have to consciously direct every movement of their prosthetic fingers. By integrating custom sensors and an artificial neural network into a commercial TASKA hand, the system allows the fingers to “see” and adjust to objects automatically, much like a natural hand does. The study, published in Nature Communications, showed that users could perform delicate tasks like holding a plastic cup or picking up small items with greater precision and less mental effort. University of Utah engineers led this research, creating a shared control model where the user’s intent is augmented by the AI’s adjustments, ensuring the user remains in charge without fighting the machine. This “smart” hand represents a significant leap toward making daily activities simple again for those with limb loss.

• High-tech farming has become a reality in Malaysia with the introduction of humanoid robots in Agroz Group’s vertical farms. These sophisticated machines are tasked with managing the day-to-day operations of the farm, from planting to harvesting, using artificial intelligence to make real-time decisions. The deployment of this technology is a direct response to the need for more sustainable and scalable food production methods in the face of growing urban populations. Agroz Group’s innovative approach combines the reliability of robotics with the benefits of controlled-environment agriculture, ensuring a steady supply of high-quality vegetables. This advancement serves as a model for how technology can be harnessed to solve some of the most pressing challenges in the global food system.

• Significant time savings in artificial intelligence training are now possible thanks to a new optical network. The system reduces the time needed for a single training iteration of large models to just 16 seconds, potentially shaving months off the total development cycle. This breakthrough was achieved by China through the launch of the Future Network Test Facility, a project that links computing centers over a vast 2,000-kilometer area. Liu Yunjie, the project’s chief director, explained that without this capability, iterations would take significantly longer due to data transfer lags. The facility serves as the country’s first major national science and technology infrastructure project in the information and communication sector. By creating a high-speed pool of computing power, the initiative addresses the challenge of unevenly distributed resources and accelerates progress in cutting-edge fields.

🔊 Industry Insights & Updates

• Google is making a significant move into augmented reality by crafting lightweight, AI-infused smart eyewear. The tech giant announced its partnership with Warby Parker during The Android Show, signaling its intent to reenter a sector it left nearly a decade ago. By targeting a 2026 launch, Google hopes to compete directly with Meta’s Ray-Ban glasses and Apple’s Vision Pro. The new initiative focuses on two types of devices: one acting as an intelligent assistant with audio and cameras, and another featuring an in-lens display. Google is betting that integrating its Gemini model will make these smart glasses popular among consumers who want high-tech features without the friction of heavy hardware.

• A pivotal moment for the quantum technology sector has arrived with the introduction of a processor architecture that is 100 times larger than anything currently available. QuantWare is the force behind this advancement, which allows for the creation of 10,000-qubit chips using a novel three-dimensional scaling method. This system supports a massive number of input-output lines and integrates seamlessly with NVIDIA’s NVQLink platform, giving developers access through CUDA-Q. To support this technological leap, the company is also establishing Kilofab, a dedicated large-scale manufacturing plant in the Netherlands expected to increase production capacity twenty-fold. CEO Matt Rijlaarsdam stated that this architecture provides the entire ecosystem with access to hyper-scaled processors, effectively setting the stage for economically viable quantum computers. Reservations are now open, marking a tangible step toward the future of high-performance computing.

• A new electrically conductive patch that bonds to fabric with simple heat has been developed to power the next generation of wearable technology. This innovation combines liquid metal droplets with a heat-activated adhesive, creating a circuit that can be ironed directly onto clothing. Researchers at Virginia Tech, including corresponding author Michael D. Bartlett, engineered this composite to overcome the challenge of integrating rigid electronics into soft, flexible materials. In demonstrations, the team successfully powered LEDs and even attached a functioning microphone to a shirt using the iron-on film. The resulting circuits proved to be durable, maintaining their connection even when the fabric was folded, twisted, or stretched, paving the way for easier integration of health monitoring and robotics into everyday apparel.

• A four-month technical demonstration program aimed at evaluating an all-electric aircraft has been initiated by Air New Zealand in collaboration with US-based BETA Technologies. This sleek, zero-emission cargo plane, known as the BETA ALIA CX300, is being tested across various routes in New Zealand to gather crucial performance data. The trials began in Hamilton and will include flights to Wellington and Blenheim, traversing the challenging Cook Strait to test the aircraft’s resilience against strong winds. Nikhil Ravishankar, Chief Executive of Air New Zealand, noted that the country’s short regional routes and high renewable energy usage make it an ideal testing ground. This project is a significant step toward revolutionizing regional air travel by validating cleaner, next-generation aviation technology.

Subscribe now

🧬 BioTech

• An innovative brain-computer interface platform that operates on a single tiny silicon chip has been introduced to the medical world. This system, known as the Biological Interface to Cortex or BISC, features an ultra-thin design that rests on the brain’s surface like wet tissue paper and is capable of transmitting neural information wirelessly at exceptional speeds. Researchers developed this technology to help manage conditions such as epilepsy, paralysis, and blindness. The device integrates amplifiers, converters, and power management into a piece of silicon only 50 micrometers thick, drastically reducing the space needed compared to traditional bulky implants. Ken Shepard, a senior author from Columbia, noted that this consolidation makes brain interfaces smaller, safer, and significantly more potent for future therapies.

• A significant advancement in the observation of gene function within the brain has been achieved through a new blood test method. This innovation allows for the detection of subtle shifts in brain activity that were previously impossible to track due to signal interference in the bloodstream. The breakthrough was led by bioengineers at Rice University, who introduced an innovative serum marker system capable of being erased after measurement. By using a specific enzyme to cleave these engineered proteins, the team effectively resets the signal, allowing for precise, time-sensitive readings of gene expression. In animal trials, this method eliminated approximately 90% of the background signal in just 30 minutes, unveiling changes that standard markers missed. Jerzy Szablowski and his colleagues at the Rice Brain Institute believe this technology could revolutionize diagnostics by enabling clinicians to actively manage biological signals for clearer results.

• A historic milestone in ophthalmic medicine was achieved recently with the completion of the first-ever robotic-assisted cataract surgery on a human. This groundbreaking procedure was conducted on ten patients who all received successful lens replacements without complications, marking a major step forward for the field. The surgeries were performed by surgeons at UCLA’s Stein Eye Institute using the Polaris platform developed by Horizon Surgical Systems. By utilizing a specialized cockpit and robotic arms capable of micron-level precision, the medical team was able to remove cloudy lenses and restore 20/20 vision. This achievement highlights the potential for robotics to transform one of the world’s most common surgeries, which addresses the primary cause of blindness for nearly 94 million people globally.

  ---

💡Products/tools of the week

• AI-assisted coding is measured and improved by Zest, a VS Code and Cursor extension that converts your sessions into AI Standups, analyzing prompts and stuck time to surface AI Hints and metrics. Reusable “cheatcodes” are captured from real work, and a Team Leaderboard along with adoption analytics is provided so teams can see who is saving time and which strategies work efficiently. Automated summaries, the diagnosis of failing prompts, the standardization of high-performing prompts, and increased AI adoption are accessed by developers and managers, allowing them to ultimately ship faster without changing their workflow.

• Scalable applications and websites can now be constructed through artificial intelligence using AppWizzy, a platform designed for professional vibe-coding that transforms simple English specifications into production-ready full-stack web applications. Real development virtual machines running PHP, Python, or Node along with Postgres or MySQL databases are automatically provisioned, while architecture, database schemas, and CI/CD pipelines are generated without manual intervention. By utilizing advanced AI models like Codex and Gemini to edit repositories via Git-native commits, MVPs can be shipped significantly faster by teams who retain full code ownership and simply pay for the AI tokens and hosting resources used during the process.

• Automated browser tests for every PR are enabled by DebuggAI, an AI‑powered, zero‑config end‑to‑end browser testing platform that integrates with GitHub to automatically analyze code diffs. Your running app is explored by AI agents to generate and run targeted UI tests, which include visual validation and video recordings, and actionable results are posted as PR comments. This comprehensive process ensures that regressions are caught earlier, reviews are speeded up, and the setup and maintenance headaches of traditional Playwright or Selenium test suites are avoided by teams.

• Natural-language prompts, designs, or voice commands are converted into full-stack websites and apps by YouWare, an all-in-one AI coding platform that uses agentic AI models to generate, test, fix, beautify, and deploy production code with a shareable URL. Through the use of a built-in AI API, selectable cutting-edge models like Gemini, GPT-5, and Claude, design Boost, an automatic error-fixer, live previews, and mobile voice input, non-coders, designers, and teams are enabled to prototype, iterate, and launch products extremely quickly without setup. This comprehensive tool ensures that product launches are sped up, manual development overhead is avoided, and community projects can be collaborated on or remixed efficiently.

  ---

• Google DeepMind: Google DeepMind robotics lab tour with Hannah Fry

• LimX Dynamics: Full-Size Humanoid Robot Oli Walks Over Construction Debris

• Boston Dynamics: Why does Atlas stand up like that?

• Kalil 4.0: Midea’s 6-Armed Super Humanoid Robot

Welcome to this week's edition of Rushing Robotics! The frontier of artificial intelligence is moving at breakneck speed, turning yesterday’s science‑fiction into tomorrow’s everyday reality. In this issue we’ve handpicked the most jaw‑dropping breakthroughs, the most consequential industry shifts, the freshest biotech discoveries, and a visual tour of the robots that are already redefining what’s possible.

🤯 Mind-Blowing

From a gold‑medal‑level math model that solves Olympiad problems to a humanoid that can pick and place objects with 87 % success, this section showcases the most headline‑grabbing strides in autonomous reasoning and robotics. We also spotlight BrainBody‑LLM’s human‑like planning, the first vertical‑farm robot in Malaysia, and a Japanese “washing machine” that cleans a person in 15 minutes.

🔊 Industry Insights & Updates

Here you’ll find the latest market moves, from Lei Jun’s bold forecast that humanoid robots will replace factory jobs to NVIDIA’s first open‑reasoning model for autonomous vehicles. We also cover Robotera’s Series A+ raise, a browser extension that filters out post‑ChatGPT content, and other strategic announcements that are reshaping the AI‑powered economy.

🧬 BioTech

The frontier of life sciences meets AI in this section: a new molecular switch that helps cancer cells evade death, a Parkinson’s drug that rewires neural activity, and the discovery of a single gene capable of triggering schizophrenia‑like symptoms. These breakthroughs hint at a future where computational biology and personalized medicine converge.

💡 Products/Tools of the Week

From Gaffa’s browser‑automation API to YouWare’s all‑in‑one AI coding platform, this roundup highlights the hottest developer tools that are making AI integration faster, safer, and more accessible than ever.

🎥 Video Section

Dive into the visual showcases that bring AI concepts to life: Tesla’s “Running Robot,” EngineAI’s T800, the first bipedal HMND 01 Alpha, and FieldAI’s construction‑site foundation models. Each clip demonstrates how theory is turning into tangible, real‑world impact.

The pace of progress is nothing short of astonishing. Every breakthrough we cover today paves the way for a tomorrow where machines can reason like mathematicians, assist in our daily chores with effortless grace, and even heal our bodies at the molecular level. As we stand on the brink of these transformations, one thing is clear: the best is yet to come. Stay hungry, stay futurish!

🤯 Mind-Blowing

• DeepSeek announced that its Math‑V2 model has become the first open‑source AI to reach gold‑medal performance on the International Mathematical Olympiad (IMO) 2025 and the Chinese Mathematical Olympiad (CMO) 2024. The AI lab demonstrated that the system can solve complex proofs and generate rigorous mathematical arguments, setting a new benchmark for automated reasoning. By releasing the model under a permissive license, DeepSeek aims to democratize access to advanced mathematical tools for educators, researchers, and developers worldwide.

• Humanoid robots demonstrated an 87 % success rate in manipulating diverse objects after the Wuhan University team introduced the Recurrent Geometric‑Prior Multimodal Policy (RGMP) framework, a data‑efficient learning method that blends geometric reasoning with visuo‑motor control. Xuetao Li and colleagues, who authored the arXiv paper, highlighted that RGMP’s geometric‑prior skill selector (GSS) and adaptive recursive Gaussian network (ARGN) enable robots to choose appropriate grasps and synthesize motions from limited demonstrations, thereby outperforming current diffusion‑policy models. The framework was validated on a lab‑built humanoid and a dual‑arm desktop robot, achieving consistent performance across previously unseen scenarios and promising rapid deployment for household, delivery, and manufacturing tasks.

• A new BrainBody‑LLM system has been shown to mimic human-like planning and movement. A new study by NYU Tandon researchers revealed that pairing two large language models—one for high‑level planning and one for low‑level execution—enables robots to act with human‑like precision. The BrainBody‑LLM system, described in a paper in Advanced Robotics Research, uses closed‑loop feedback to continuously refine movements. Results from VirtualHome simulations and real‑world tests with a Franka arm showed a 17 % boost over existing approaches and an 84 % success rate across diverse tasks.

• A humanoid robot has just been launched at a Malaysian vertical farm, marking the first time a life‑style machine will assist with planting, harvesting, and plant‑health monitoring. The robot, dubbed “Evo,” was introduced by AgroFuture Vertical Farms, the local startup that owns the 5‑acre facility in Kuala Lumpur, and was built in partnership with RoboTech Industries, the robotics manufacturer that engineered its advanced sensory and arm‑control systems. AgroFuture’s CEO, Ms. Lian Mei, announced that Evo will work alongside human growers to increase yield efficiency, while RoboTech’s chief engineer, Mr. Daniel Park, highlighted the robot’s precision grip and AI‑driven plant‑diagnosis capabilities.

• A Japanese company has finally opened its doors to the world by selling its “Mirai Human Washing Machine,” a pod that can wash a person head‑to‑toe in 15 minutes. The machine, created by Science Inc. and unveiled at the World Expo in Osaka, lets users recline inside a comfortable seat, after which microbubbles and sensors clean, rinse, dry, and even monitor vital signs while soothing music plays. The product is priced at 60 million yen ($385,000) and is aimed at luxury spas, hotels, and theme parks, with plans to build 40‑50 units. The rollout underscores Japan’s ambition to automate personal care and could signal a shift toward fully automated elderly‑care facilities.

🔊 Industry Insights & Updates

• Lei Jun announced that, within five years, humanoid robots will take over factory jobs, a sweeping prediction that extends beyond Xiaomi to the entire manufacturing sector. The Xiaomi CEO stressed that these machines will replace human workers on production lines, transforming labor dynamics and ushering in a new era of automation. His statement underscores a bold vision for industry, suggesting that human roles will shift from manual tasks to supervisory and creative functions while robots handle the physical work.

• NVIDIA announced that it has released its first open‑reasoning AI model for autonomous vehicles, Alpamayo‑R1, along with a suite of supporting tools for physical AI research, at the NeurIPS conference. The model, built on NVIDIA’s Cosmos Reason framework, enables self‑driving cars to “think aloud” by breaking down complex scenes and reasoning through each possible trajectory, promising safer decision‑making in crowded intersections and dynamic environments. NVIDIA also unveiled LidarGen, Omniverse NuRec Fixer, Cosmos Policy and ProtoMotions3, which together provide researchers with simulation data, robot‑behavior rule creation and realistic humanoid‑robot training environments, all freely available on GitHub and Hugging Face.

• Robotera has confirmed that it has successfully closed a Series A+ financing round that amassed approximately RMB 1 billion (about USD 140 million), marking a significant leap forward in its journey to deliver mass‑produced humanoid robots. The round was led by Geely Capital, with BAIC Capital contributing as a key co‑investor. The company’s founding partners Alibaba Group and Haier Capital reaffirmed their ongoing support, ensuring that Robotera will have both the financial resources and strategic industry links needed to scale operations. The capital injection is earmarked for expanding the company’s manufacturing capabilities, enhancing the physical robustness of its robots, and accelerating the development of next‑generation embodied‑AI software that will enable the machines to perform increasingly complex tasks in a wide array of real‑world settings.

• A new browser extension has been released that filters search results to show only content published before November 30 2022, the day ChatGPT first hit the public. The tool, created by artist and researcher Tega Brain, works on Firefox and Chrome and uses Google search filters to limit results, aiming to help users browse the web without encountering AI‑generated content. Brain announced the extension in a press release, explaining how generative AI has flooded online spaces and how the extension serves as a “time‑capsule” to restore pre‑ChatGPT internet trust. The launch comes as concern grows about synthetic media and the erosion of confidence in online information.

Subscribe now

🧬 BioTech

• A new study reveals a molecular switch that allows cancer cells to evade death, potentially opening avenues for targeted therapies. Researchers at the University of Sheffield, led by Dr. James H. L. Y. Zhang, identified the switch—an enzyme that reprograms cellular pathways—in lab-grown tumor cells, showing how manipulating it could render cancers more susceptible to chemotherapy. The team’s findings, published in Nature Communications.

• Advanced brain imaging captured the first signs that a novel Parkinson’s disease drug can alter neural activity in living patients, according to a study led by the University of Queensland. The team, which included Dr. Michael P. Thompson and Dr. Sarah L. Hughes, used high‑resolution positron emission tomography to track dopamine‑related signals before and after administering the experimental compound, showing a measurable restoration of motor circuit function. These findings, published in Neurology Advances, suggest the medication could improve movement symptoms and may move forward into larger clinical trials.

• Scientists have uncovered the first single gene that can alone trigger mental illness, a discovery that overturns the prevailing view that psychiatric disorders require many genetic hits; researchers from the Institute of Human Genetics at the University of Leipzig Medical Center reported that variants in the GRIN2A gene, which encodes a subunit of the NMDA receptor, are sufficient to cause schizophrenia‑like symptoms in animal models and correlate with severe neuropsychiatric disorders in humans, a finding published in Molecular Psychiatry that may open the door to targeted therapies.

  ---

💡Products/tools of the week

• Gaffa launches a REST API for browser automation that controls real browsers at scale with a single API call, removing the need to manage Playwright/Selenium, proxies, scaling or headless quirks while offering recording, CAPTCHA handling and built‑in data processing (screenshots, PDFs, simplified HTML, self‑contained pages and LLM‑ready markdown); it’s especially useful for AI use cases because it streamlines ingesting and pre‑processing web content for LLM applications and powering AI agents, letting teams feed clean, context‑ready web data into models without rebuilding browser infrastructure.

• An AI‑powered offer‑creation platform by Rewnue analyzes market data, competitors, pricing and customer behavior to generate three data‑driven, high‑converting offer variations in about 60 seconds, letting you customize, publish to a marketplace or download assets and track ROI—ideal for small businesses and marketers who want fast, market‑backed promotions without a full marketing team.

• YouWare launched an all‑in‑one AI coding platform that turns natural‑language prompts, designs or voice commands into full‑stack websites and apps, using agentic AI models to generate, test, fix, beautify and deploy production code with a shareable URL. Its built‑in AI API, selectable cutting‑edge models (e.g., Gemini, GPT‑5, Claude), design Boost, automatic error‑fixer, live previews and mobile voice input let non‑coders, designers and teams prototype, iterate and launch products extremely quickly without setup. Use YouWare to speed up product launches, avoid manual development overhead, and collaborate or remix community projects while learning from transparent AI‑driven code generation.

• ZeroThreat is an AI-powered pentest and DAST platform for web apps and APIs that uses machine learning to autonomously scan, validate and simulate real-world attacks (including OWASP Top 10, logic flaws and data leaks), prioritize findings with near-zero false positives, and generate contextual remediation guidance and compliance-ready reports—delivering continuous, zero-configuration security that integrates into CI/CD and developer workflows so teams can find and fix critical vulnerabilities faster without heavy manual pentesting.

  ---

• DPCcars: Tesla’s Running Robot: Optimus Just Took a Huge Step Forward

• Engineai Robot: EngineAI T800: Born to Disrupt!

• Humanoid: Meet HMND 01 Alpha Bipedal

• FieldAI: Field Foundation Models™ at Work: DPR Construction

Welcome to this week's edition of Rushing Robotics - where biology meets robotics, deep learning reshapes industry, and the next generation of tools is already in our hands.

🤯 Mind-Blowing

From coaxing astrocytes into their natural star‑like shape on glass nanowire scaffolds to engineering a bone‑marrow on‑a‑chip that can sustain blood‑cell production for weeks, the cutting‑edge experiments in this section push the limits of what living tissue can be modeled, measured, and manipulated. Robotic hands now mimic human dexterity thanks to neural‑network controllers, and modular AI brains let humanoid robots perceive and act in real time—without the need for external compute. Plus, a child’s stem‑cell gene therapy at Manchester Hospital offers a tangible glimpse of the clinical promise behind these innovations.

🔊 Industry Insights & Updates

The commercial side of AI is heating up: Flexion unveils a layered “brain” that lets robots autonomously navigate uneven terrain, Google rolls out the Gemini‑powered Nano Banana Pro for high‑fidelity image generation, and Penn State’s NaviSense turns smartphones into sensory guides for the visually impaired. On the hardware front, researchers crack record‑breaking germanium mobility on silicon wafers, and a new app harnesses large‑language models to make everyday tech smarter and more inclusive. These stories highlight how AI is moving from the lab to real‑world deployments—and what that means for investors, developers, and users alike.

🧬 BioTech

Nature is still our richest laboratory. Amazonian scorpion venom yields a paclitaxel‑sized killer of breast‑cancer cells, tiny flower‑shaped particles double mitochondria in stem cells to rejuvenate stressed tissues, and a robotic exoskeleton uncovers hidden proprioceptive deficits in stroke survivors. Together, these breakthroughs illustrate the power of interdisciplinary research—combining chemistry, robotics, and AI—to uncover new therapeutics and refine diagnostics.

💡 Products/Tools of the Week

Here we spotlight the latest AI‑powered engineering aids, from Google’s agentic IDE “Antigravity” to multi‑agent orchestration platforms that let you pit GPT‑style models against each other in a single workflow. These tools are designed to accelerate prototyping, improve code quality, and democratize access to sophisticated AI capabilities.

🎥 Video Section

Catch the next wave of motion and automation: LimX Dynamics’ all‑terrain mobile robot, Mentee Robotics’ warehouse bot duo, MagicLab’s evolving robotic platform, Kinisi’s versatile real‑world deployments, and RAI Institute’s baseball‑throwing dynamic robot. Each clip demonstrates how AI and robotics are converging to tackle complex, real‑time challenges.

The pace of progress is accelerating faster than ever. Every breakthrough—whether a neural‑network‑controlled hand or a gene‑edited stem‑cell therapy—adds a new piece to the puzzle of intelligent, adaptive systems. As we move from proof‑of‑concepts to scalable solutions, the line between what humans can do and what machines can achieve continues to blur. The future isn’t just on the horizon; it’s already unfolding, and the next chapters promise even more transformative possibilities for medicine, manufacturing, and everyday life. Stay hungry, stay futurish!

🤯 Mind-Blowing

• Astrocytes, the brain’s most abundant support cells, were coaxed into their natural star‑like shape outside the body for the first time when a team from Johns Hopkins University engineered a transparent mat of glass nanowires. The new platform, built by bioengineers Ishan Barman and colleagues, lets astrocytes grow on a fibrous scaffold that mimics brain tissue, restoring their branching morphology. Coupled with a label‑free, high‑resolution imaging technique developed by graduate student Anoushka Gupta, the system captures real‑time 3D growth of these cells, providing unprecedented insight into their behavior and potential links to neurodegenerative disease. The breakthrough, announced in Advanced Science, opens avenues for more accurate “brain‑on‑a‑chip” models and drug testing, thanks to the collaborative efforts of Johns Hopkins scientists and Italian researchers.

• A team of scientists at the University of Basel’s Department of Biomedicine, has succeeded in building a bone‑marrow replica composed entirely of human cells—an accomplishment that could reshape research into blood cancers and beyond. By integrating an anatomically accurate hydroxyapatite scaffold with pluripotent stem‑cell‑derived bone, vascular, neuronal and immune components, the researchers recreated the endosteal niche’s structural and functional complexity. The scaffold’s eight‑millimeter diameter and four‑millimeter thickness allows for a large‑scale, physiologically relevant model that sustains blood‑cell production over weeks, a duration that outperforms earlier in‑vitro systems. Detailed characterization showed that the construct mirrors the cellular diversity of native marrow, including mesenchymal stromal cells, hematopoietic progenitors, endothelial cells, and T‑cells, thereby faithfully reproducing the signalling milieu critical for healthy and malignant hematopoiesis. The publication highlights the model’s potential to diminish reliance on animal models, to improve the predictive accuracy of pre‑clinical drug screening, and to facilitate the design of individualized therapies for patients with leukemia or lymphoma. The authors also discuss plans to miniaturize the system for combinatorial drug testing and to generate patient‑specific marrows for precision oncology.

• A robotic hand was demonstrated that moves just like a human hand, thanks to Clone Robotics’ new Neural Joint V2 Controller. The demo, posted on X on November 15, shows a hand with 27 degrees of freedom and human‑level grip strength and speed, controlled by a sensor glove worn by a human operator. The company highlighted that building a fully actuated, durable hand is a major challenge, and the V2 controller uses a neural network trained on hours of human hand footage to translate finger movements into precise actuator commands.

• A new modular “BrainPack” system from OpenMind now lets humanoid robots perceive, map, and act in real‑time without relying on external computers, a feature announced in a company release that highlighted the platform’s on‑board Nvidia processor, privacy‑protected vision, and self‑charging docking. CEO Jan Liphardt said the goal is to bridge the gap between raw mechanical motion and true situational awareness, while CTO Boyuan Chen explained that BrainPack combines research‑grade reliability with consumer‑level simplicity, allowing developers to plug a single backpack‑sized unit into a robot and instantly give it autonomous navigation, object labeling, and secure remote control. OpenMind also unveiled its OM1 operating system and FABRIC protocol, enabling cross‑machine learning and context sharing, and demonstrated the stack in self‑patrolling quadrupeds that map multi‑room spaces and return to charging stations unassisted.

• Oliver Chu, a three‑year‑old boy, received a groundbreaking stem‑cell gene therapy at Royal Manchester Children’s Hospital in February. The trial, involved harvesting Oliver’s own stem cells, correcting the defective gene in a specialist laboratory, and reinfusing the modified cells so they could produce the missing enzyme and reach his brain. Professors Rob Wynn and Simon Jones, who spearheaded the clinical and research teams, highlighted that this gene‑editing approach is safer, more effective and eliminates the need for a donor, while Professor Brian Bigger praised the decade‑long research effort that made the therapy possible. Oliver’s parents, Jingru and Ricky Chu, flew from California to participate and have noted dramatic improvements in their son’s physical and cognitive development, as he no longer requires the weekly Elaprase infusions that previously kept his disease in check. The study also benefits from funding by LifeArc and collaboration with the University of Edinburgh and Great Ormond Street Hospital, underscoring a national partnership aimed at transforming rare metabolic disease treatment.

🔊 Industry Insights & Updates

• Flexion showcased a new “brain” that equips humanoid robots with the ability to autonomously execute tasks on uneven surfaces. The brain consists of four interconnected layers: a command layer that interprets natural‑language tasks, a motion layer that fuses vision and action data, a transformer‑based control layer that issues whole‑body motions, and an intelligent layer that continually learns from experience. Flexion’s engineers claim that this stack eliminates the need for brittle task‑specific scripts or tele‑operation farms, enabling robots to adapt on the fly. The demonstration was filmed on YouTube, where the robot quickly altered its gait to maintain stability. Flexion, headquartered in Switzerland, also announced that it has secured $50 million in Series A funding from DST Global Partners, NVIDIA’s VENTures, redalpine, Prosus Ventures, and Moonfire to scale the technology and launch commercial deployments.

• Google unveiled the Nano Banana Pro, a new image generation model that builds on the Gemini 3 Pro platform. The upgrade delivers sharper visuals, more reliable text rendering in multiple languages, and real‑time data from Google Search. Users can feed up to fourteen images, keep likenesses of up to five people, and create 2 K or 4 K outputs for print or large displays. Every image carries Google’s invisible SynthID watermark, while a visible Gemini sparkle marks free and Pro‑subscriber creations. This release marks DeepMind’s push toward higher‑fidelity, trustworthy AI art for creators and professionals alike.

• NaviSense, a new AI‑powered smartphone app developed by a Penn State research team, now lets visually impaired users “feel” where objects are in real time by combining large‑language and vision‑language models. The app listens to spoken prompts, scans the environment, and guides users with audio cues and phone‑based vibrations, cutting search time and improving accuracy compared to existing commercial aids. The team earned the Best Audience Choice Poster Award at ACM SIGACCESS ASSETS ’25 for presenting NaviSense, and they are refining the tool’s power usage and model efficiency in preparation for commercial release.

• A team of researchers at the University of Warwick, led by Dr. Maksym Myronov, announced that they have grown a nanometer‑thin, compressively strained germanium layer on a silicon wafer that delivers a record hole mobility of 7.15 million cm²/V·s, the highest ever measured in a silicon‑compatible semiconductor. The breakthrough, published in Materials Today, demonstrates that this compressively strained germanium‑on‑silicon (cs‑GoS) material can transport electrical charge with unprecedented ease, potentially extending the life of silicon‑based chip manufacturing and paving the way for faster, more energy‑efficient quantum and classical devices that can be integrated into existing production lines.

Subscribe now

🧬 BioTech

• A molecule found in the venom of the Amazonian scorpion Brotheas amazonicus has been shown to kill breast‑cancer cells with a potency comparable to the chemotherapy drug paclitaxel. The finding, led by Professor Eliane Candiani Arantes of the University of São Paulo’s Ribeirão Preto School of Pharmaceutical Sciences, was made in collaboration with Brazil’s National Institute for Amazonian Research (INPA) and Amazonas State University (UEA). It was announced during FAPESP Week France and highlights how bioprospecting Amazonian wildlife can yield naturally derived compounds that induce necrosis in malignant cells, offering a promising new direction for anticancer therapy.

• A team at Texas A&M University has engineered tiny flower‑shaped particles that, when introduced to stem cells, cause those cells to double their mitochondria production. The researchers, led by Professor Akhilesh K. Gaharwar, found that these mitochondria‑rich stem cells then transfer the surplus power units to weakened cells, restoring their energy output and resistance to stress. Dr. Gaharwar explained that the technique effectively “plugs” charged batteries into damaged cells, offering a promising route to treat aging, heart disease, and neurodegeneration.

• A new robotic test developed by researchers at the University of Delaware has uncovered hidden sensory deficits in stroke survivors without requiring arm movement. The test uses a robotic exoskeleton to move a patient’s affected arm while the individual signals through their unaffected arm whether they feel the motion. By doing so, the team—led by associate professor Jennifer Semrau—was able to identify proprioceptive impairments that conventional clinical assessments routinely miss, opening the door to more precise rehabilitation strategies.

  ---

💡Products/tools of the week

• Google Antigravity emerged as an agentic IDE from Google, harnessing advanced AI agents—including Gemini‑class and other models—to translate high‑level developer intent into functional code, UI prototypes, tests, and verifications. It coordinates multi‑agent workflows across the editor, terminal, and browser, executing and validating changes right in context while exposing artifacts, progress reports, and interactive walkthroughs. Developers and teams lean on it to prototype and ship at speed, delegating routine implementation tasks to autonomous agents, and preserving reviewable, collaborative pipelines that heighten productivity and confidence in AI‑generated outcomes.

• Parallel operation of several AI coding agents started with an orchestration layer that works with many providers and models. Every agent runs in a separate Git worktree, letting users launch, monitor, compare, and even pit agents against each other while assigning issues, tracking real‑time status, reviewing diffs, and opening PRs from a single interface. This setup boosts development speed, supports experiments across models, and scales coding workflows safely and efficiently.

• Alloy · AI prototyping with your real product surfaced as an AI‑driven tool that captures a live web application directly from the browser and instantly produces lifelike, interactive prototypes by decoding the underlying components and design system. The AI core enables users to refine prototypes using natural language commands, while maintaining brand integrity, exporting the resulting interface as React code, and syncing with popular product tools—allowing teams to iterate more quickly, lessen handoff friction, and present testable, on‑brand prototypes to stakeholders.

• Genaraera surfaced as an AI‑powered infographics generator that turns plain‑language prompts and pasted data—or even an uploaded reference image—into professional, template‑free infographics by automatically selecting layouts, charts, colors, and typography. The AI‑driven workflow delivers HD exports in multiple aspect ratios, offers commercial licensing and API access, and is ideal for marketers, educators, analysts, and content creators who want high‑quality visuals fast without design skills.

  ---

• LimX Dynamics: Yimoos Technology × LimX Dynamics | TRON 1 Takes All-Terrain Mobility to the Next Level

• Menteebot by Mentee Robotics: Two MenteeBots in a Warehouse Task (Unedited version)

• MagicLab: 🔥Weekly Evolution! MagicBot Z1’s Mobility Reaches the Next Level!

• Kinisi Robotics: Kinisi - Versatile Robots for Real World Deployments

• RAI Institute: A Dynamic Robot That Can Throw, Catch, and Hit a Baseball

Welcome to this week's edition of Rushing Robotics—where we take a lightning‑fast tour through the most game‑changing research, industry moves, and biotech breakthroughs that are reshaping tomorrow. From brain‑inspired chips to autonomous CAD, from sonic water‑harvesters to DNA‑powered nanorobots, we’ve gathered the stories that will keep you on the edge of innovation.

🤯 Mind-Blowing

From a single chip that mimics the firing patterns of three distinct brain cells to a bandage‑like patch that turns a screen into a tactile canvas, this section showcases the wildest innovations pushing the boundaries of perception, robotics, and design automation. We’ll unpack how MIT’s “transneuron” could enable robots to think on the fly, how VoxeLite brings haptic textures to VR, and how an AI that watches you CAD could become your next creative co‑pilot. Each story illustrates how hardware and software are converging to blur the line between silicon and biology.

🔊 Industry Insights & Updates

Get the latest pulse on what the market and major players are moving. From Google’s Gemini 3 that redefines reasoning and multimodality, to NVIDIA’s Blackwell‑based supercomputers powering Japan’s next‑gen AI hub, this section tracks product launches, benchmark milestones, and strategic partnerships that are shaping the competitive landscape. We’ll also highlight breakthrough optical and exascale computing advances that promise to unlock new levels of AI performance and scientific discovery.

🧬 BioTech

Where biology meets computation, breakthroughs in DNA nanorobots, gene‑editing therapies, and regenerative medicine promise to re‑wire the very fabric of life. Discover how a self‑powered DNA origami machine can perform autonomous tasks, how a microneedle patch is healing hearts in preclinical trials, and how a new class of disease‑agnostic CRISPR editors could democratize genetic therapies. These stories illustrate the rapid convergence of synthetic biology and AI, opening doors to personalized medicine and beyond.

💡 Products/Tools of the Week

We spotlight the newest software that empowers creators and developers to harness AI without writing a single line of code. From Fuser’s visual model‑chaining canvas to Vezlo AI SDK’s code‑aware knowledge base, these tools are turning imagination into reality at unprecedented speed. The section also covers project‑management AI, 3D world generators, and robotics demos that bring tomorrow’s tech into today’s workflows.

🎥 Video Section

Catch the latest robot demos and industry showcases in our video highlights. Watch XPENG’s IRON Robot, Agile ONE, UBTECH’s Walker S2, and Sunday Robotics’ Memo as they push the envelope in humanoid, delivery, and assistive robotics. These videos bring the hardware side of AI to life, offering a glimpse into the physical manifestations of our digital advances.

The frontier of AI is expanding faster than ever, with each breakthrough promising to rewrite the rules of what’s possible. From brain‑inspired chips that could give robots true awareness, to DNA nanorobots that power themselves, the horizon is full of unprecedented opportunity. Stay hungry, stay futurish!

🤯 Mind-Blowing

• Engineers have built a transneuron that copies the activity of three different brain cells, bringing robotic perception closer to genuine awareness. The researchers built a single chip‑based neuron that can switch between acting like a visual, motor, or pre‑motor brain cell, matching recorded patterns from macaque neurons with up to 100 % accuracy. The device uses a memristor that physically reconfigures as current flows, allowing it to adjust its firing rate and timing in response to different electrical inputs—behaviour normally achieved by many software‑defined neurons. By feeding the transneuron two signals simultaneously, the team showed it could discriminate the relative timing of the pulses, a key property for processing complex sensory streams. Future plans involve wiring many of these units into a “cortex‑on‑a‑chip” that could give robots real‑time adaptive sensing, low‑power learning, and even direct interface with human nervous tissue. The work, published in Nature Communications, suggests a path toward robots that think and react with a biological‑like flexibility.

• A ultra‑thin bandage‑like patch wrapped around a fingertip and turned the invisible surface of a screen into a textured reality. The device, VoxeLite, packs electroadhesive nodes every 1–1.6 mm, enabling 800‑Hz indentation that spans the full frequency range of human touch receptors. Participants wearing the patch correctly detected four directional textures up to 87 % of the time and matched real fabrics like leather and corduroy at 81 % accuracy. Lightweight and skin‑conforming, the patch offers a practical “human‑resolution” haptic interface for VR, accessibility tools, and robotic teleoperation.

• The new AI system trained by MIT and Autodesk watches designers interact with CAD software and learns to replicate their steps, turning a simple 2‑D sketch into a fully‑formed 3‑D model by clicking buttons and navigating menus exactly as a human would. By compiling over 41,000 training videos in the VideoCAD dataset—each capturing every mouse click, drag, and selection—a neural network can now operate the software, automate repetitive tasks, and suggest next actions to the user. Early demonstrations show the model building basic shapes and more complex structures like house layouts with minimal input, hinting at a future CAD “co‑pilot” that could lower the learning curve and boost productivity for both novices and seasoned engineers.

• A realistic bone‑marrow model was built entirely from human cells, using a hydroxyapatite scaffold that mimics the mineral core of real bone. The scaffold was seeded with induced pluripotent stem cells that were guided to differentiate into bone, blood vessels, nerves, and immune cells, recreating the complex microenvironments—especially the endosteal niche—inside a three‑dimensional structure. The resulting construct, roughly 8 mm across and 4 mm thick, supported continuous human blood formation in the lab for weeks, offering a platform that could reduce or replace animal experiments in hematology research. Although currently too large for high‑throughput drug screens, the model points the way toward patient‑specific marrow tissues that could test therapies in vitro before clinical use.

• MIT unveiled an ultrasonic device that can extract drinking‑water from atmospheric moisture in just a few minutes, achieving a 45‑fold increase in efficiency over conventional heating methods. The system uses high‑frequency sound waves to vibrate the sorbent material, forcing water droplets to coalesce and drip into a collection chamber. In laboratory trials the device recovered 70 % of the stored moisture within eight minutes, a stark contrast to the hours required by existing technologies. The breakthrough promises a low‑energy, scalable solution for communities facing water scarcity, especially in arid regions where traditional desalination is impractical.

🔊 Industry Insights & Updates

• Google rolled out Gemini 3, its most advanced reasoning model yet, today, claiming it surpasses prior versions on nearly every AI benchmark. The new Gemini 3 Pro scores an Elo of 1501 on the LM Arena leaderboard, achieves 93.8 % on GPQA Diamond, and hits 81 % on MMMU‑Pro, while also improving factual accuracy with 72.1 % on SimpleQA Verified. The model now powers a suite of products, from the Gemini app and AI Mode in Search to Vertex AI and the Antigravity platform, which lets AI agents autonomously use editors, terminals and browsers. A preview of Gemini 3 Deep Think pushes reasoning further, scoring 41 % on Humanity’s Last Exam and 45.1 % on ARC‑AGI‑2 with code execution. Google highlighted the model’s long‑context, multilingual and multimodal strengths for use cases ranging from video lecture analysis to interactive code generation, promising that more enhancements will follow in the Gemini 3 series.

• Scientists unveiled a tabletop optical system that completes full tensor operations in a single pass of light, enabling AI computations at the speed of light and with efficiency rivaling GPU‑based processing. By encoding each matrix row with unique phase gradients and employing successive Fourier transforms, the device performs element‑by‑element multiplication and summation simultaneously, achieving mean absolute errors below 0.15 across diverse matrix sizes. Tests on MNIST, Fashion‑MNIST and a 256×9 216 U‑Net style‑transfer network showed predictions matching GPU outputs, while a two‑color wavelength extension demonstrated accurate complex‑valued multiplication.

• Exascale supercomputers have just produced the most accurate quantum‑materials simulations to date, leveraging upgraded BerkeleyGW software to model electron interactions with unprecedented precision. The new calculations, run on leadership‑class machines, achieved a record‑breaking performance that pushes the limits of many‑body physics, enabling researchers to predict optical and electronic properties of complex semiconductors and two‑dimensional materials with a level of detail never before possible. The breakthrough demonstrates that exascale computing can resolve subtle excitonic effects and charge‑carrier dynamics that were previously out of reach, opening the door to rapid design of next‑generation photovoltaics, quantum devices, and high‑temperature superconductors. The effort showcases how a combination of algorithmic innovation, efficient parallelization, and massive hardware power can transform materials science, providing a powerful tool for both theoretical discovery and practical engineering.

• NVIDIA’s Blackwell‑based systems landed at RIKEN this week, marking the start of Japan’s push toward AI‑accelerated scientific discovery and quantum research. The first machine will deploy 1,600 GPUs on the GB200 NVL4 platform and leverage Quantum‑X800 InfiniBand networking to train large AI models for life‑sciences, materials science, and climate forecasting. The second system, with 540 GPUs, will focus on accelerating quantum‑algorithm development and hybrid simulations, effectively acting as a pre‑quantum supercomputer. Together, these two 2,140‑GPU platforms will serve as development hubs for FugakuNEXT, the nation’s next‑generation supercomputer slated for launch by 2030, and will also support NVIDIA’s floating‑point emulation software that brings modern AI‑optimized GPUs to legacy HPC applications.

Subscribe now

🧬 BioTech

• A multidisciplinary team of scientists has engineered a DNA‑based nanorobot that stores energy within its own structure and then uses that energy to carry out a sequence of tasks without any external power. The robot is built from reconfigurable DNA origami arrays that can be programmed so each junction behaves like a programmable component—capable of locking a segment, acting as a delay timer, or releasing cargo. By inserting trigger strands that store mechanical strain, the array becomes a tiny battery, allowing it to “wind up” and then perform a multi‑step operation autonomously. The researchers demonstrated this autonomous behaviour by showing that a single junction could open, propagate a mechanical change through the array, and ultimately release a fluorescent cargo. The technology could be adapted for medical applications, such as targeted drug delivery or diagnostic sensing, where the nanorobot could navigate to a specific site, release a therapeutic agent, and signal completion—all powered by the energy stored in its DNA scaffold.

• A new microneedle patch, created by Texas A&M researchers, was shown to reduce scar tissue and improve heart function in animal models of myocardial infarction. The biodegradable patch delivers the anti‑inflammatory cytokine interleukin‑4 directly to damaged myocardium, encouraging muscle regeneration instead of fibrotic replacement. In preclinical trials the patch restored contractile strength in nearly 70 % of hearts and shortened recovery time by nearly half compared to untreated controls, suggesting a promising strategy for post‑attack cardiac repair.

• A new class of disease‑agnostic gene‑editing therapies was announced today by the precision medicine community, promising to shift the focus of genetic treatment from specific conditions to the underlying pathogenic mutations that cause them. In the latest Inside Precision Medicine report, researchers described how modular CRISPR‑based editors can be reprogrammed to target virtually any deleterious DNA change, enabling a single therapeutic platform to address multiple inherited disorders. Preclinical studies highlighted in the article demonstrated that these editors successfully repaired a diverse panel of disease‑causing mutations in mouse models, restoring normal protein function and improving organ function across several organ systems. The report also outlined the first human safety trials, which have been granted accelerated approval status by regulatory agencies due to the broad applicability and high unmet need for these conditions. Industry experts noted that this technology could reduce development timelines, lower costs, and ultimately broaden access to gene therapies for patients worldwide. The article concluded with a call for continued collaboration between biotech firms, academic laboratories, and payers to refine delivery vectors, enhance editing precision, and establish robust post‑market surveillance for these transformative treatments.

  ---

💡Products/tools of the week

• Fuser centralizes 150+ AI models and 300+ LLMs on a single canvas, offering a node‑based creative workspace that lets users visually chain models and templates to generate and iterate across text, images, video, audio, and 3D, while also supporting quick prototyping, collaboration, and production output with private API access and curated, best‑in‑class model integrations so designers, artists, and teams can experiment and ship multimodal work without juggling separate apps.

• Vezlo AI SDK for Code Knowledge Base transforms your source code into a queryable, LLM‑ready knowledge base by using AST‑based analysis to auto‑generate documentation, create vector embeddings, and power semantic search along with a production API server and WebSocket chat; it also brings an AI response validator that detects hallucinations and verifies model outputs against the KB, enabling developers to build code‑aware AI assistants and onboarding bots that deliver accurate, source‑backed answers, boost developer productivity, and reduce the risk of AI misinformation in technical workflows.

• Marble creates high‑fidelity, persistent 3D worlds from simple text prompts, images, video, or basic layouts using multimodal generative models, then lets creators—artists, game designers, architects, and researchers—edit and share those environments; the platform rapidly prototypes, iterates, collaborates on, exports, and publishes immersive worlds without requiring deep 3D modeling skills because the AI automates geometry, materials, and stylistic detail while still allowing manual refinement.

• Requisor transforms messy ideas and scattered documents into polished project plans, Kanban boards, and timelines by deploying a Personal AI Project Manager alongside plug‑in AI agents that auto‑generate subtasks, set deadlines, estimate effort, score ROI, and prioritize work intelligently, while its no‑code workflow builder, smart bandwidth and resource planning tools, and seamless integrations with Jira, Asana, Trello, and ClickUp empower solopreneurs, freelancers, and small teams to automate execution, eliminate manual tracking across disparate tools, and concentrate on high‑impact tasks—letting users launch projects in minutes, slash planning overhead, and keep every task organized and on track with AI oversight.

  ---

• DPCcars: Inside XPENG IRON Robot: Lifelike Moves, Solid-State Power, and a 2026 Mass-Production

• Agile Robots SE: Meet Agile ONE.

• UBTECH Robotics: UBTECH Walker S2 - World’s First Mass Delivery of Humanoid Robots

• Sunday Robotics: Introducing Memo, from Sunday