It's official! OpenMind is hosting a Smart Fleet Meet & Greet with Bay Area Robotics Association on October 6th 5PM - 8PM during SF Tech Week @Techweek_ Come meet industry leaders in physical AI and our smart fleet powered by OM1. RSVP below via Partiful.

Phase Action Space for cross-embodiment Representation (PHASOR for short) introduces a universal action representation that maps different humanoid embodiments in a shared motion space, making robot behavior more transferable, interpretable, and scalable across platforms. At OpenMind, our deployments involve mixed fleets that operate on one brain. PHASOR helps our team run shared interpretable actions that are transferable across devices without a need for additional re-training. PHASOR is a collaboration with @AIMSafety_, @LGE_Global, Seoul National University and MAUM AI. Read the research here:

When's the ChatGPT moment for Physical AI? @DrJimFan's physical Turing test is an excellent one, but focuses on the ability of machines to reproduce human performance and behavior. In many areas, physical AI already significantly exceeds human capabilities. How do we also capture things like the extreme safety and reliability of a Waymo or the ability of some humanoids to outrun Usain Bolt? Let's broaden what we expect of advanced physical AI, to include: 1/ Rapid learning of extreme physical dexterity 2/ Capable of diverse cognitive, planning, and physical tasks 3/ Inherent safety through constitutional robotics The details and rationale:

OpenMind joined our partner TRMI in celebrating 40 years of innovation and brought along one of our humanoid robots for the occasion. Congratulations to the entire TRMI team on this incredible milestone. We’re proud to celebrate with you and excited for what’s ahead.

Robots should not care about your lighting. Our paper on Domain-Invariant Latent Lookahead (DILL) was selected for a presentation at RSS 2026. VLA policies quietly learn visual shortcuts: change the camera angle, the background, or the light, and they execute the wrong task. DILL teaches the policy to predict a domain-invariant future, not the pixels. This is the work behind the promise: OpenMind builds robot experiences that stay reliable in the real world, where the lights, the layout, and the cameras never match the lab. Huge thanks to our partners at @SeoulNatlUni, @Hyundai, Ajou University, and Tommoro Robotics. Stay tuned for more from OpenMind Research.

What would you do if you could borrow a robot body for a while? Instabody lets you drive a real humanoid robot and see the world through its eyes. We built it because the best way to learn what robots should do is to hand people the controls. Waitlist is open at

Introducing OpenMind's storefront for general-purpose robots. Whether you’re exploring automation for work, research, or everyday use, BotPop makes it easier to find the right robot for the job. Need help deciding which robot form factor works best for your environment or use case? Reach out to support@botpop.com for a consultation.

We drove our robots from San Francisco to New York City. Along the way we presented at Automate in Chicago, the biggest robotics convention in North America, took a ferry so our robot could see the Statue of Liberty, and wrapped up the trip with NYC’s first robot pop-up shop in SoHo. Thousands of people visited. Most had never met a robot before, and now they have. Thanks to @KraneShares and our partners: @nvidia, @Visa, @AGIBOT_US, @UnitreeRobotics, @boosterobotics, @LimX_Dynamics, @GalaxeaDynamics, @intbotai, and @ARCL_Katena.

Our Unitree Go2s are getting better at navigating real-world clutter. In this demo, it can now self-detour from corners and avoid obstacles using adjusted MPPI parameters. Fully autonomous navigation means being able to move without bumping into potentially critical objects. In workplaces and homes this is especially important.

Our build night was a huge success! Thank you to all the technical teams for showing up and launching on OM1. Winners received exclusive OpenMind backpacks and showcased their amazing demos including: - Robot that mirrors human motion in real time - New robot form-factor abstracted and automated in just 30 minutes - Autonomous drones controlled in simulation by natural language It will only get easier to build applications from here.

All roboticists - in case you are building VLAs, please stop what you are doing, take a break, get a coffee, and read the LeWorldModel paper by @lucasmaes_, @randall_balestr, @ylecun and collaborators: And then read it several more times - the same general approach can be directly mapped to other key problems in robotics, including dealing with multimodal inputs such as vision and speech

OpenMind OM1 Build Night w/ @OpenAI Codex Location: San Francisco [Address given upon successful RSVP] Date: Wednesday, May 6 @ 4:30 PM - 9:00 PM ​This event is for robotics and agent developers, AI-native builders, technical founders, and curious engineers who want a practical way to learn OM1 by actually building with it. ​Bring a laptop and come ready to ship something. Register:

We were delighted to participate at CONNECT 2026: Global Embodied AI Innovation Summit where we formalized our strategic partnership with @MagicLab_Robot. We also shared our thoughts in the panel: Perception, Foundation Models & Decision-Making. Excited to see continued collaboration in the space as we expand our reach across multiple robot manufacturers.

Why Generic Humanoid Robots Will Fail — And What's Next Imagine an alternate world where we never invented the car. In that world, a robotics engineer might reasonably conclude that robotic horses are the future — replace the living ones, keep the stables and saddles, ride them to work. Convenient, modern, and the roads stay free of manure. It sounds absurd only because you already know about cars. We keep making the same mistake with humanoid robots. Consider transportation. To finally make driving safe, we had two options: put a humanoid in the driver's seat, or embed sensing and compute directly into the vehicle. Waymo chose the latter. It has no steering wheel. It exists purely to move people efficiently from A to B. The humanoid was not needed. Consider a sock factory. Yes, you could replace workers with humanoid robots one-for-one on the assembly line — and gain maybe 2-3x efficiency. Or you could completely redesign the workflow around a purpose-built autonomous sewing system and eliminate most of the factory, the chairs, the cafeteria, the manual sewing machines, the HVAC, the doors, and the restrooms. The actual optimization is to side-step the previous human-imposed physical constraint. Look at Ukraine. The front lines aren't filling up with Terminator-style humanoids carrying rifles. Human soldiers are being replaced by heterogeneous swarms of purpose-specific drones: some for reconnaissance, some for logistics, some for delivering munitions. War is being restructured around the desired outcome (survival), not the soldier's shape. Consider a 1970's office. Want to move information through teams of people? We once used typists, paper, trucks to supply the paper, typewriters, and repair technicians. A linear improvement would have been to replace the human typist with a 10-fingered humanoid. What actually happened? The entire workflow — paper, printers, typewriter factories, delivery trucks, the desks, the offices — was obliterated. Email deleted the human clerk's entire universe. Consider cancer early detection by mammography. Today, getting a mammogram requires expensive hardware, logistics infrastructure, human nurses and doctors, a biopsy workflow, a human pathologist with a microscope (imported from Germany or Japan), a written finding, multiple physician reviews. Sure, you could replace the pathologist with a humanoid (the microscope focus knob requires finger dexterity) and get a modest efficiency gain (and faster responses at 2 am). Or — the far more likely future — we all swallow a cancer detection pill every few months, and 24 hours later a color-changing sticker on our arm turns red or green. No hardware. No hospital. No logistics. No pathologist. No office. No desk. No humanoid. The workflow isn't optimized by a literal drop-in swap of a human pathologist for a humanoid. The entire workflow simply ceases to exist. Consider life sciences research and drug development. We're seeing excitement about robot arms and humanoids pipetting water in research labs. Robot horses, episode 7. We don't design aircraft by crashing test planes — we simulate them entirely in software first. Biology will go the same way. The path to scalable drug discovery isn't robot arms in conventional wet labs demonstrating 10 fingered prowess in manipulating Eppendorf tubes filled with purple food coloring. Rather, we need in-silico biological models that evaluate billions of hypotheses computationally, with physical manipulation of atoms only at the very end. The clear pattern. Efficient automation doesn't try to replicate a 10-fingered human in a static context. Automation eliminates physical rate-limiting steps in their entirety. That's why "classical" humanoid robots, as a generic category, will largely fail. They're robotic horses. They assume the infrastructure and workflows stay fixed and only the 10-fingered human is swapped out. That's not how economic and technological pressure works. What actually matters? If humans continue to inhabit the physical world, then moving atoms will remain important, and that requires five things: atoms, energy, force generation and actuation, sensing, and compute. Everything else — form factor, number of limbs, type of end effector — is a variable to be optimized for the task. So if you are a pathologist, a robotics engineer, a teacher, a parent, a politician, or a sewing factory owner - please think different. Most obviously, we should all anticipate, and build for, a future in which robots exhibit extreme physical fluidity: Two arms or four. Wheels or legs. Tentacles or flippers. Three fingers or twelve, or none at all. Eyes at the front, side, or tip of a tentacle. At OpenMind, we don't care what you look like right now - we got you, in all your physical form factors. OM2 ships in July, for all machines. Let's build.

This week, we were featured in Forbes and The Robot Report where our Founder, @JanLiphardt gave his thoughts on: Physical AI’s Future - Microsoft-OpenAI Alliance Overhauled As Battle For AI Dominance Widens - Take a look at how OpenMind is navigating the changing AI + robotics landscape.

From gesture to intent, our latest work shows how advanced keypoint detection (body + hands) can unlock powerful, real-time action recognition. We’re pushing the boundaries with both data-driven models and zero-shot approaches, scaling from a handful of core actions to a richer set of human behaviors without always needing new training data. This is a glimpse into more adaptive, intelligent systems that understand people naturally and is vital for mass robot adoption.

Demonstrating our latest localization system in action. At bootup, the robot has no prior position estimate, but within seconds it autonomously localizes itself using a fusion of three algorithms. Unlike our previous version, this updated system incorporates vision, allowing the robot to adapt to real-world changes (e.g., moved furniture) by recognizing previously seen environments. In this demo, we repeatedly reset navigation and reposition the robot to random locations, showing robust, repeatable localization. The robot then executes a full patrol, following a planned path (visualized in RViz) with real-time path tracking. Because our software is hardware-agnostic, it brings the same reliable performance to any robot it runs on.

2/ Unitree Robotics @UnitreeRobotics In its third Gala appearance, Unitree brought robotic martial arts to the stage, performing alongside professional athletes. Precision. Balance. Control. A clear signal of hardware maturity and advanced motion intelligence- the prerequisites for reliable, large-scale real-world adoption.