I will defend my PhD thesis, “Physics-Constrained Generative Models for Computational Design,” on Tuesday, September 8, 2026, at PM ET. Location: MIT @MIT_CSAIL 32-G449 Zoom: mit.zoom.us/j/93799161698 Committee: Kaiming He, Bill Freeman, Wojciech Matusik Generative models can now propose candidate designs at high throughput, yet the physical world remains the ultimate judge: a candidate becomes a solution only if it can be realized in the real world and satisfies the functional requirements of the design problem. (1/7)

Making human-robot interaction easy.

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: janliphardt.com/2026/08/26/a…

Excited to see @Reuters cover the launch of our startup Accelerated Understanding. We are training large scale AI models that can simulate and understand physics to invent and discover. Our models understand the world directly in 4D (3D + time) and across physical phenomena. Going full 4D requires massive context length, we have pushed it to a Trillion in training and exceeding 5 Trillion at inference. AI giving you a bigger haystack of ideas doesn’t help. The bottleneck for new inventions and discoveries is shifting from ideas to the ability to test them. With AI that can simulate and understand physics we are directly attacking this bottleneck. People have been trying to do this for a while now, but usually by taking shortcuts. Narrow surrogates are great if you happen to have enough of precisely the right data and your design loop stays in distribution. Video models look fantastic but sweep physical accuracy under the rug, and some static world models cut out physics altogether. A lot of interesting physics isn’t visual. What does not cutting corners look like? Space stays 3D and you also have time: so 4D in total. You also need multiple physical modalities in the same model, not just things you can see. That’s what we’ve built. Scaling is the primary ingredient to make this work. To represent the world you need sufficient context, which in our case grows in 4 dimensions. Individual samples get so big they don’t fit into single accelerators or even full nodes anymore. We’ve developed architectural tricks to make it work. We’ve pushed our models to 1T parameters during large scale pre-training and are able to train at up to a Trillion context when needed and do inference exceeding 5 Trillion context without any sub-sampling or patching. Building on prior successes of AI weather forecasting, fusion simulation, design of medical devices, drugs and chips, we wanted to see if scale and universality can benefit AI for physical understanding. With our teams’ experience in large-scale infrastructure and model training we’ve been able to pull it off. reuters.com/business/ai-foun… acceleratedunderstanding.com… @accelerated_u @bjenik

Nature research paper: Human brain organoids record the passage of time over multiple years go.nature.com/4gnGNV4

Mathematics in the age of AI by Terence Tao arxiv.org/abs/2608.16753

Physics is going to be as cooked/cooking as math. I fed Claude an open problem in stochastic thermodynamics of the kind I'd suggest to a mathematically inclined grad student. And over a few days of back and forth, it did months of work and closed the whole problem class.

I'm a former Citadel quant who covered power & gas. There's constant talk about chips & memory, but power is the central bottleneck for AI. Very few people understand it, so I'm posting a canonical primer on power pricing & data centers: power2026.ai/

One runs on kibble, the other runs on code. 📸: @KraneShares

Today marks the final day of our NYC pop up shop! We had thousands of visitors and millions of impressions generated as attendees shared pictures and videos of our robots. It was a pleasure collaborating on this with @KraneShares and our other partners: @nvidia, @Visa, @AGIBOT_US, @UnitreeRobotics, @boosterobotics, @LimX_Dynamics, @GalaxeaDynamics, @intbotai, and @ARCL_Katena.

OpenMind showcased 4+ robot form factors at Automate Chicago — unstructured social scenarios, public interaction. Founder @JanLiphardt: "robots must be safe for humanity." Trust as the foundation.

Chicago Automate was a huge success! For many attendees, this was their first time interacting with a single robot let alone many different robots in social scenarios. We let loose 4+ robot form factors where people could interact in completely unstructured scenarios. Founder @JanLiphardt also shared his thoughts on how robots must be safe for humanity, and that only though building foundational trust can we create lasting solutions. We're continuing to bring robots across the world to show what we mean at OpenMind.

Reporting live from @AutomateShow in Chicago! 🦾 Come learn about the future of socially-intelligent machines at our booth in the @NVIDIA Humanoid Pavilion in McCormick South Hall (Booth #2091).

Very excited to welcome @NoamShazeer to OpenAI as our new lead for architecture research! His work on transformers, MoE, and efficient decoding have shaped modern AI. He’s extremely AGI-pilled and is super thoughtful about making it all go well. Welcome, Noam!

OpenMind and @KraneShares are hosting New York City’s first pop-up store selling general-purpose robots. From June 26–28, visitors can experience the world’s most socially-intelligent robots up close, all integrated with our OM1 software. Customers will be able to place pre-orders via OpenMind's new online store for all things robotics. Take a glimpse into the future of intelligent machines at BotPop: 188 Lafayette St, SoHo, New York City 10 AM–7 PM daily! Learn more about our store below:

Most AI investing happens downstream of the frontier: a capability emerges, a category gets named, and capital rushes in. But by the time a category earns a clean box on a market map, the best builders have usually been living in the messy version for months. Agents. Reasoning. RL environments. World models. AI for Science. Recursive self-improvement. I call this frontier proximity: the ability to see what is becoming possible before it becomes consensus. My frontier proximity ladder: L0 Wrapper: uses today’s models. L1 Reactor: reacts fast to releases, but roadmap is downstream. L2 Anticipator: builds for where capabilities are going. L3 Native: depends on a non-obvious frontier bet. L4 Shaper: helps move the frontier itself. The point is not that every company needs to train models. Apps can have high frontier proximity if they understand what models will make possible next. Infra can have high frontier proximity if it knows what future agents, multimodal systems, robotics stacks, or scientific workflows will need. That is why we’re launching MoE Capital. MoE stands for Mixture of Experts. The idea is simple: build an AI fund around people closest to the frontier: frontier researchers, technical founders, AI-native builders, and seasoned operators. We don’t want to be another AI fund with a newsletter-level understanding of the frontier. We want to build the AI fund closest to the frontier. More in The Information: theinformation.com/newslette…

Robots nodding for coordination might seem unnecessary. Why don't they communicate through internal messaging? The case for maximally independent agents: Mixed-Fleet Environment: Humanoids must collaborate with humans and robots from diverse manufacturers that lack shared private protocols. Social Legibility: Human-centric communication make intent readable, building trust and helping them fit well into social environments.

Bots, Bevs & Devs was a blast! Huge thanks to everyone who came out to Circuit Launch in Oakland for an evening of robotics, AI demos, short talks, and great conversations. OpenMind had the chance to demo our Unitree B2 robot dog running OM1, and it was awesome seeing people interact with the future of embodied intelligence up close.