DeepMind's Geoffrey Irving: AI companies could just stop unilaterally

ShakeelHashim · x · 2026-09-29

Geoffrey Irving of DeepMind's frontier safety team said in an interview: "If you're doing a very dangerous thing, you should just slow down." He argues AI companies overplay the framing of safety as a pure coordination problem — they could unilaterally stop dangerous development without waiting for everyone else. A notably blunt take from a major lab insider.

Original post →

More from AGI Musings

AGI Musings channel →