Richard Ngo: How AI safety got captured by OpenAI and later Anthropic
RichardMCNgo · x · 2026-09-11
Former OpenAI researcher Richard Ngo traces how the AI safety agenda was captured first by OpenAI and later Anthropic, promising follow-ups on OpenAI's internal power dynamics during his 2021-2024 tenure.
On Paul Christiano joining OpenAI, he says he cut an outright condemnation from an earlier draft: Christiano remains one of the few morally serious thinkers oriented to superintelligence risk, but his strategic judgment is badly mistaken and the community should discount it accordingly.
Related event: Ex-OpenAI Researcher: AI Safety Agenda Co-opted by OpenAI and Anthropic(3 posts)→
More from AGI Musings
- Anthropic's S-1 Shows $6.6M Revenue Per Employee as AI Agents Log 6x Human Hours — jacalulu · 2026-10-26
- Ben Recht: The world needs a truly open language model, and it's easier to build than you think — beenwrekt · 2026-10-26
- Which jobs fall to AI first? The rule: any layer whose rules can be written as code — FinanceYF5 · 2026-10-26
- Anthropic S-1: expert-level AI 'within a few years', 90% of code written by Claude — kimmonismus · 2026-10-26
- The identity crisis in software engineering: multiagent systems demand a new problem set — rakyll · 2026-10-26
- Frontier LLMs negotiate contracts poorly and frequently renege for selfish gain, study finds — xuanalogue · 2026-10-26