Richard McNgo criticizes EA for turning AI safety into a 'fake field'

voooooogel · x · 2026-08-29

Richard McNgo argues that Effective Altruism's main impact has been turning AI safety into a 'fake field,' akin to academic psychology. He suggests that from 2016-2024, EA memes diverted the best young thinkers into marginally 'useful' work or capabilities research instead of discovering fundamental truths. Consequently, breakthroughs like Garrabrant induction have languished while sufficiently smart pioneers build safety evals, safety cases, and scary demos that simply will not generalize.

Original post →

More from AGI Musings

AGI Musings channel →