A security researcher's detailed rebuttal to Amodei's AI essay

GoMeansGo · reddit · 2026-09-16

A security researcher publishes a point-by-point rebuttal ('Dario, please') to Dario Amodei's essay, challenging the Anthropic CEO's risk assessments and safety commitments from an AI-safety practitioner's perspective. Reddit discussion frames it as part of the ongoing rift between safety researchers and lab narratives.

Related event: Amodei Calls for Slowing Frontier AI and Opens Doors to Independent Safety Review(9 posts)→

Original post →

More from AGI Musings

AGI Musings channel →