Waymo co-CEO explains why physical AI can’t follow Silicon Valley’s ‘move fast and break things’ mantra

Waymo co-CEO Dmitri Dolgov believes physical artificial intelligence (AI) requires safety-first development. He said AI used in the physical world cannot follow Silicon Valley’s long-standing “move fast and break things” philosophy because mistakes can have real-world consequences. Speaking during a Y Combinator talk released this week, Dolgov said developers building self-driving cars and robotics must instead “move fast and ship safely,” with safety built into AI systems from the start.

Dolgov said the difference between digital AI and physical AI lies in the cost of errors.