AI race is not a Prisoner's Dilemma: Is it rational to stop capabilities research?

geoffreyirving · x · 2026-08-08

Geoffrey Irving shares a viewpoint that the AI race is not a Prisoner's Dilemma. He argues that it is not rational for anyone to be doing capabilities research at a frontier lab right now. The situation is very dangerous, and if one person or lab stops, it makes it easier and more peer-compatible for others to stop.

Related event: AI Race Is Not a Prisoner's Dilemma: Unilateral Pause on Capabilities Is Rational(5 posts)→

Original post →

More from AGI Musings

AGI Musings channel →