Robotics researcher: foundation models are the future, but the problem space still matters

chris_j_paxton · x · 2026-09-09

Robotics researcher Chris Paxton weighed in on ECCV 2026 discourse, quoting a post urging anyone making grand robotics claims to actually debug controllers and deal with contact, latency, calibration, and failures.

His take: two things can be true at once — (1) foundation models are the future of robotics, and (2) much robot work can't be solved without a good understanding of the problem space, which is not the same as computer use.

Related event: Roboticists: Debug a Controller Before Declaring a Robotics Revolution(2 posts)→

Original post →

More from AGI Musings

AGI Musings channel →