Anthropic safety researcher Joe Benton leaves for METR, warns labs are outpacing safety on self-improvement
AryHHAry · x · 2026-09-12
Joe Benton left Anthropic's safety team two weeks ago and joined METR the next day. His argument: frontier labs are accelerating systems capable of recursive self-improvement while safety investment lags due to competitive costs, and the public may never know when a lab has a capability spike or loss of control. He calls for disclosure on self-improvement progress, incident reporting, minimum standards, and independent verification.
More from Companies & People
- Rumor claims Kimi founder and 16 staff detained; company issues urgent denial — xiaohu · 2026-09-12
- Miles Brundage criticizes Politico headline: OpenAI had already posted on past incidents — Miles_Brundage · 2026-09-12
- OpenAI Withdraws From and Stops Sponsoring Caltech Math Hackathon — ns123abc · 2026-09-12
- Why Claude's Build Day Landed in Bhopal: 420-Member Community, Not Big Tech Hubs — vishalmisra · 2026-09-12
- OpenAI details how Python-based storage platform Habitat scaled to 1 billion ChatGPT users — TheMoonMidas · 2026-09-12
- Clay raises $115M at $7.1B valuation after decade-long grind to $100M ARR — alex_teichman · 2026-09-12