In 2018, a simulated humanoid learned to do a backflip by watching a motion capture clip. In 2025, the same lineage of research produced a single model that controls a real Unitree G1 robot with zero task-specific training. The path from one to the other redefines what ’learning to move’ means for robots, virtual humans, and the film industry.
Cathie Wood sees a golden age. Ray Dalio sees a bubble. BlackRock is quietly tokenizing everything. Meanwhile, robots just started clocking in at BMW. Three industries are merging into one economic stack, and no one agrees on what happens next. This is an attempt to make sense of it.
In a Google DeepMind lab, a robot arm folds paper into a fox shape from fewer than 100 demonstrations. At ICLR 2026, 164 papers compete to define robot intelligence. From RT-1 to BitVLA, the science behind Physical AI is moving faster than the machines themselves.
Jensen Huang calls it the ChatGPT moment for robotics. Boston Dynamics just started shipping Atlas to Hyundai. A Chinese startup sells humanoid robots for less than a used Honda Civic. Physical AI, the idea that machines can perceive, reason about, and act in the real world, went from conference slide to factory floor in under two years. Here is what that actually means.
Behavior Foundation Models can make virtual characters backflip and real humanoids walk. But making them help a stroke patient climb stairs requires solving problems the robotics community has barely started to name. Here is where the technology actually stands, what is missing, and which research directions might close the gap.
A CMU roboticist says the hardware is solved. The real challenge is building a brain that understands your body. From experiment-free RL to neural interfaces, the science of wearable Physical AI is converging, and so is the market.