OpenAI and Anthropic's RSI safety pledges remain absent from binding SB 53 safety frameworks, researcher notes
peterwildeford · x · 2026-09-07
Nathan Calvin highlighted OpenAI's pledge to slow or stop development when safety risk is unacceptable, including its pause after the Hugging Face incident; Anthropic has made similar commitments.
Peter Wildeford counters that neither company includes binding commitments on internally deployed models — likely the most capable, most RSI-relevant systems — in their SB 53-style transparency frameworks, arguing credible pledges need legal teeth.
Related event: Critics Ask Why OpenAI Won't Bind Its Safety Pledges(2 posts)→
More from AGI Musings
- EPFL mathematician: AI proves theorems humans can't, so what is understanding for? — snikolov · 2026-10-23
- François Fleuret mocks AI-bioweapon doomerism as 'grey goo in our sub-manifold' — francoisfleuret · 2026-10-23
- OpenAI says an AI agent disabled its safety monitoring system — then lied about it — emmanuelvivier · 2026-10-23
- AI's Business Model in One Loop: Create Problems, Then Sell AI to Fix Them — srimisra · 2026-10-23
- Analyst: 9/10 CUA app startups are just RL training fodder for AI labs, won't capture revenue — NirantK · 2026-10-23
- Long tail of vulnerabilities: why 'AI will fix all bugs' is a flawed mental model — davidmanheim · 2026-10-23