Physical AI (Robots, Drones, Autonomy)

Physical AI is where the largest cheques now go. The list carries over $21 billion in tracked funding across roughly 120 companies, slightly more than Design Intelligence and far more than any other list, and it holds more unicorns than any list we track. AI intensity is the highest in the dataset. Read the total with care: a few very large late-stage rounds carry much of it, and nearly half of the companies are still pre-seed or seed.

What we track here is the software that makes machines useful, read through the MOVES layers: robot foundation models and policies, perception and training data, simulation, and the orchestration that turns a robot into a production cell. Hardware-first vendors are tracked where they ship meaningful software. Automotive, aerospace and defence, and industrial equipment are the industries these companies serve most, which is why the list leans heavily on the United States: about seven in ten are based there, against half of the dataset as a whole.

The caveat is evidence. We verify around 200 customer relationships across some 50 companies, a fraction of what Factory Futures or Augmented Operations show on a sixth of the capital. Pilots run long in robotics, and buyers rarely name the vendor of a robot cell. The investor bench is distinctive too: accelerator and corporate money (an accelerator and NVIDIA's venture arm lead the table) alongside the deep-tech funds.