Researcher's Guide to Physical AI

Shubham Shrivastava
  • 5.0 (2)
  • Technology
  • Updated Weekly

Every week, a deep-yet-accessible 15-minute briefing on what happened in Physical AI: robot foundation models, humanoids, manipulation, simulation, world models, and the research behind them. Built for researchers and curious newcomers alike, every episode gives you the background you need, then goes deep with concrete examples. Each episode ships with a long-form written deep dive. Every episode ships with a long-form written deep dive, full method details, benchmark numbers and links to every paper, at https://physicalaipodcast.co

Episodes

  1. 4d ago

    Anthropic Aims for a Record IPO, Unitree Up 460%

    Anthropic has told its bankers it expects the largest first-time share sale in history, and it is reportedly still trying to buy a world-model company for six billion dollars while it does. A private lab can fund a five-year bet on physics, and a public one answers to a quarterly clock, which makes this the most consequential open question in Physical AI right now. Also: Unitree closed its first day of trading up 460 percent on a shipment number three organisations can't agree on within fifty percent, three labs independently decide the headroom is in the loop around the policy rather than in the weights, and a new benchmark finds that up to a quarter of recorded successes on soft objects crushed the object. Full write-up for this episode, every number, method detail and citation, at https://physicalaipodcast.co/episodes/003/ More episodes and the complete archive: https://physicalaipodcast.co

  2. Aug 15

    Your Simulator Has Never Seen a Robot Fail

    About twenty researchers spent a paper writing down, formally, the rule that when you tell a learned simulator to move the robot's arm, the arm should move. Six of the open-source world models the field builds on don't reliably obey, and a second paper explains why: trained only on successful demonstrations, they've never had a reason to learn what a bad action does, so feed them one and they show you the task succeeding anyway. Also: Google DeepMind's whole-body humanoid stack, which contains no world model at all, a robot that triples its success rate under a lighting change because it learned where to look, and the training examples that poison a dataset by being correct. Full write-up for this episode, every number, method detail and citation, at https://physicalaipodcast.co/episodes/002/ More episodes and the complete archive: https://physicalaipodcast.co

  3. Aug 8

    Looks Like Physics, Isn't Physics, and World Action Models

    Two teams tested, independently and in the same week, whether the video-generating world models robotics is betting on actually understand physics. They get the shape of a falling mug right and the numbers underneath it wrong, which matters, because these are the simulators meant to replace real robot data. Also: why choosing your training video beats collecting more of it, a robot that spots a fake instruction and obeys it anyway, and a short primer on world action models, including why imagining the future turns out to matter in training but not at run time. Full write-up for this episode, every number, method detail and citation, at https://physicalaipodcast.co/episodes/001/ More episodes and the complete archive: https://physicalaipodcast.co

  • 3 Episodes

Ratings & Reviews

5
out of 5
2 Ratings

About

Every week, a deep-yet-accessible 15-minute briefing on what happened in Physical AI: robot foundation models, humanoids, manipulation, simulation, world models, and the research behind them. Built for researchers and curious newcomers alike, every episode gives you the background you need, then goes deep with concrete examples. Each episode ships with a long-form written deep dive. Every episode ships with a long-form written deep dive, full method details, benchmark numbers and links to every paper, at https://physicalaipodcast.co

Information