Weekly Piece of Future #169
From Intent-Reading Brain Implants to Vision-Restoring Prostheses and Touch-Sensing Surgical Bots
Hey there, fellow future-addicts!
Welcome to this week's edition of Rushing Robotics! As we cross the midpoint of May 2026, the pace of breakthroughs shows no sign of slowing—from brain implants restoring vision to blind participants, to robots working full eight-hour autonomous shifts, the future isn't just arriving—it's accelerating. Grab your favorite beverage and settle in, because what follows will reshape how you think about what's possible.
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
Prepare to have your perceptions stretched. This section delivers the kind of breakthroughs that make you pause and reread—twice. From wireless brain implants bypassing the eyes entirely to restore artificial vision, to a higher-level brain-computer interface that interprets intent rather than just movement, the line between science fiction and science fact continues to blur. Add in rice-sized optical sensors giving surgical robots a sense of touch, humanoid robots completing full shifts with unified AI, and researchers shattering a 40-year-old atom manipulation record, and you have a lineup that redefines what machines and humans can achieve together.
🔊 Industry Insights & Updates
The deeper currents driving technology forward deserve attention. Here we explore the developments reshaping industries and research paradigms—from a quantum material that regains superconductivity under crushing magnetic fields, to a geometry-based quantum swap gate achieving 99.1 percent precision across 17,000 qubit pairs. We also cover ultrathin solar cells turning everyday glass into power generators and a major deployment agreement putting thousands of wheeled humanoid robots into global manufacturing facilities by 2032. These aren't just incremental improvements—they're foundational shifts.
🧬 BioTech
Biology meets engineering at the cellular level. This week brings a Japanese laboratory where robots—not humans—now conduct medical experiments, a jellyfish-inspired soft robot breaking speed records while navigating tight anatomical spaces, and the isolation of a rare boron peroxide molecule under remarkably mild conditions that could transform pharmaceutical production and carbon capture. The automation of discovery itself is underway, and the implications for healthcare, manufacturing, and environmental science are staggering.
💡 Products/Tools of the Week
Every week brings fresh tools to supercharge your workflow, and this lineup is no exception. From Atomic Chat's fully local, agent-enabled LLM experience that runs entirely offline with no accounts required, to Innogath's ability to transform a single question into a fully cited, multi-chapter report with auto-generated diagrams—these are tools designed for serious productivity. SynthBoard.ai simulates an entire boardroom of specialized AI advisors that debate and challenge assumptions rather than politely agreeing, while KeyAPI gives developers a single REST key to access AI-ready social data from over 20 platforms. Whether you're a developer, researcher, founder, or executive, there's something here to elevate how you work.
🎥 Video Section
Seeing is believing. This week's visual showcases feature Unitree's GD01 manned transformable mecha, a novel approach to versatile humanoid manipulation through "touch dreaming," and RLDX-1—the dexterity-first foundation model designed specifically for robot hands. Watch the future unfold in motion.
We stand at an inflection point where the convergence of neuroscience, quantum physics, robotics, and biotechnology is creating possibilities that would have seemed fantastical just a decade ago. Each breakthrough in this newsletter isn't an isolated achievement—it's a building block in a larger architecture of human capability expansion. The most exciting part? We're still in the early chapters of this transformation, and the pace is only accelerating. The question is no longer whether the future will be extraordinary, but how quickly we'll adapt to live within it Stay hungry, stay futurish!
🤯 Mind-Blowing
A successful implantation of a wireless brain implant has restored artificial vision signals in a third blind participant. Illinois Institute of Technology led the research, while Dr. Sepehr Sani performed the surgery at Rush University Medical Center. The Intracortical Visual Prosthesis bypasses the eyes entirely, using 34 wireless stimulators with 544 electrodes to send electrical pulses directly to the visual cortex. Following a recovery period, the participant will train at The Chicago Lighthouse to learn how to interpret these signals for navigation and basic tasks. Researchers will monitor the system's long-term safety and usability over the next one to three years.
A world-first higher-level brain-computer interface has been successfully implanted in a paralyzed patient to interpret intent rather than just muscle movement. UCHealth and the University of Colorado Anschutz Medical Campus performed the surgery on Brandon Patterson, who has been paralyzed from the chest down since a 2017 accident. Unlike standard devices that target the primary motor cortex, this implant taps into higher-functioning brain areas responsible for planning and decision-making. Shortly after the procedure, Patterson experienced phantom sensory feedback, feeling his paralyzed fingers twitch. He is now undergoing motor training to translate his thoughts into digital actions like moving a cursor, while researchers study the long-term effects of this bidirectional learning.
A rice-sized optical sensor has been developed to give surgical robots a sense of touch. Researchers at Shanghai Jiao Tong University in China created the 1.7-millimeter device that measures force, pressure, and twisting in all directions using light instead of electronics. The technology uses an optical fiber with a soft elastomer tip that deforms upon contact, changing the light pattern which is then analyzed to calculate force. This allows machines to detect unsafe contact early and adjust in real time during delicate procedures. The team demonstrated the sensor's ability to identify hidden structures like tumors in tissue models, aiming to improve safety in minimally invasive surgery.
Full eight-hour autonomous shifts have been achieved by humanoid robots using a new unified AI system. Figure AI developed the Helix-02 technology to combine vision, touch, and whole-body control into a single learning network, eliminating the need for separate controllers. The company also introduced System 0, a learned whole-body controller trained on over 1,000 hours of human motion data that replaces 109,000 lines of hand-engineered C++ code. This advancement allows the robots to perform multi-minute tasks without resets, such as unloading a dishwasher for four continuous minutes. Figure AI previously tested these machines at BMW facilities where they completed 10-hour shifts and moved over 90,000 parts.
A 40-year-old record for atom manipulation has been broken by researchers who moved tens of thousands of atoms in minutes. The team from the Massachusetts Institute of Technology (MIT) and Oak Ridge National Laboratory (ORNL) used an electron beam to create 40,000 quantum defects in just 40 minutes, a stark contrast to the 1980s IBM experiment that took hours to move 35 atoms. Julian Klein, an MIT researcher who conceived the project, explained that the technique uses very few electrons to avoid damaging the crystal while moving columns of chromium atoms. Frances Ross, a professor at MIT, noted that this method acts like a photocopier for atomic defects, allowing for the creation of stable structures in three dimensions at room temperature.
🔊 Industry Insights & Updates
An unconventional quantum material has been shown to regain its zero-resistance state under crushing magnetic fields due to internal magnetic fluctuations. Researchers at the Institute of Science and Technology Austria (ISTA) found that uranium ditelluride (UTe2) exhibits a strong transverse magnetic susceptibility that functions as electron glue. While most superconductors are destroyed by high magnetic fields, UTe2 revives at fields between 40 and 70 Tesla. Valeska Zambra, the lead author and a PhD student, and Kimberly Modic, an assistant professor, achieved this by mounting samples smaller than a grain of salt on a cantilever. By mechanically oscillating the crystal, they simulated a changing magnetic field direction to measure the susceptibility, proving that this hidden magnetic property is responsible for the reentrant superconductivity.
A breakthrough in quantum computing stability has been achieved by developing a new geometry-based quantum swap gate. Researchers at ETH Zurich created this operation to make neutral-atom qubits more reliable by exploiting the geometric phase, which depends on the overall path atoms take through a crystal of light rather than laser intensity. This method reduces errors caused by laser fluctuations that typically plague swap gates. The team demonstrated the technology on 17,000 qubit pairs, achieving 99.1 percent precision within a millisecond. They also successfully created half-swap gates that generate correlations between qubits, a necessity for running complex quantum algorithms.
A breakthrough in solar technology has resulted in the creation of ultrathin, semi-transparent solar cells that can turn everyday glass surfaces into power generators. Researchers at Nanyang Technological University (NTU) developed the perovskite cells to be roughly 10,000 times thinner than a human hair while maintaining high efficiency. The team, led by Annalisa Bruno, used a thermal evaporation process inside a vacuum chamber to create highly uniform layers as thin as 10 nanometers without toxic solvents. The semi-transparent version achieved a 7.6 percent efficiency while allowing 41 percent of visible light to pass through, making it color-neutral and suitable for windows. Opaque versions reached efficiencies between 7 and 12 percent. The university has filed a patent through NTUitive and plans to improve durability for commercial use.
A major agreement has been signed to deploy thousands of wheeled humanoid robots across global manufacturing facilities. UK robotics firm Humanoid partnered with German automotive supplier Schaeffler to integrate these systems into live operations, targeting a four-digit number of robots by 2032. The initial phase runs from December 2026 to June 2027 at two German sites, where the robots will handle boxes in Herzogenaurach and undergo testing in Schweinfurt. Operating under a Robot-as-a-Service model, Humanoid will provide the robots, fleet management, maintenance, and 24/7 support. Schaeffler will also become the preferred supplier for more than half of Humanoid's joint actuator demand through 2031, supplying a seven-digit number of actuators for the wheeled platforms.
🧬 BioTech
Medical experiments previously performed by human researchers are now being conducted by machines in a new Japanese laboratory. The Institute of Science Tokyo has launched the facility featuring ten robots, including the dual-armed Maholo LabDroid designed for delicate laboratory procedures. Operating without on-site human staff, the robots automatically carry out programmed cell cultivation and handle temperature-controlled equipment. This initiative directly addresses growing labor shortages and the need to minimize human error in repetitive experimental work. The university intends to expand the robotic workforce to approximately two thousand by twenty forty. The long-term objective is to automate nearly the entire research process, a vision shared by Insilico Medicine, which recently deployed a bipedal humanoid robot named Supervisor to advance autonomous drug discovery.
A record-breaking jellyfish-inspired soft robot has been developed to perform complex biomedical tasks using external magnetic fields. Researchers led by Professor Quanliang Cao created the device, known as the Jellyfish Magnetic Soft Robot (J-MSR), which achieves a swimming speed of 14.85 body-lengths per second. By mimicking the asymmetric contraction and relaxation of real jellyfish through an optimized trapezoidal magnetic waveform, the robot moves efficiently without onboard power or bulky buoyancy systems. The team demonstrated the robot's ability to navigate tight spaces in an ex vivo pig stomach model and carry payloads like microneedles with high targeting accuracy.
Cleaner industrial chemistry and advanced manufacturing may soon benefit from the isolation of a rare boron peroxide molecule. Researchers at MIT created the compound, known as dioxaborirane, under remarkably mild conditions at room temperature, avoiding the extreme cold or high pressure usually required for reactive oxygen structures. The molecule features a strained ring containing one boron and two oxygen atoms, functioning like a compressed spring that stores chemical energy. Lead author Chonghe Zhang emphasized that the ability to generate these compounds under mild conditions opens the door to entirely new chemical reactions. The newly discovered structure demonstrated two distinct behaviors, transferring oxygen atoms to other molecules for potential pharmaceutical production and reacting with carbon dioxide for possible capture applications.
💡Products/tools of the week
Atomic Chat introduces this fully local, agent-enabled experience as a free and open-source application. It runs large language models entirely on device, supporting over 1,000 options like Llama and Gemma via GGUF and ONNX. The integration of TurboQuant ensures faster inference and better memory management, while the built-in agent capabilities allow the AI to automate tasks and keep persistent local memory. With no accounts required and all processing happening offline, the app provides a secure, uncapped, and customizable tool for developers and privacy advocates.
Innogath transforms any question into a fully cited, multi-chapter report complete with auto-generated diagrams and an interactive visual canvas. The platform features branching research pages that keep insights accessible rather than lost in linear chat formats. Its AI-driven Deep Research pipeline ingests PDFs and web pages, cross-referencing thousands of trusted sources to ground all claims. Users can select between research and chat modes while the system auto-generates context-aware branches and exportable notebooks. The tool serves researchers, students, analysts, and founders who require fast, verifiable, and well-organized research outputs.
SynthBoard.ai simulates a boardroom of specialized AI advisors called Synths to help users navigate complex strategic, hiring, product, or personal decisions. Users drop in a brief, and the system auto-assembles or lets them choose multiple persona-matched Synths that debate, challenge assumptions, and defend positions rather than politely agreeing. Powered by multi-LLM routing and persona tuning, the platform produces consensus-scored recommendations, ranked ideas, action items, and an audit trail. It also learns from real outcomes via integrations and allows users to build custom Synths. This makes it ideal for founders, executives, and teams who want fast, auditable, multi-perspective AI advice that scales like an affordable advisory board.
KeyAPI introduces an AI-agent-first solution that gives developers a single REST key to access profiles, posts, videos, comments, analytics, and commerce signals from over 20 platforms. The system is built specifically for LLM pipelines, automation, and autonomous agents by streaming real-time and historical social data for up to 1,000 days as clean, AI-ready JSON. By removing the complexity of multiple keys and rate limits, KeyAPI allows teams to train models, enrich prompts, power agent workflows, and deploy scalable social intelligence much faster.
