Ex-Insider Questions Anthropic's Claim That "AI Safety's Technical Problems Are Solved"
ZhongRuiqi · x · 2026-10-29
ZhongRuiqi says he thought it was fake until he opened the report: Anthropic's latest safety assessment leads with the pillar "the technical problems of AI safety are solved."
The cited page lists progress on constitutional alignment, interpretability-based model audits, and misuse safeguards including biology classifiers and adversarial jailbreak testing. A former Anthropic employee who once believed the company would genuinely solve AI safety, he argues the stated limitations on page two directly contradict the pillar-one framing, and is disappointed Anthropic is highlighting it.
Related event: Anthropic's claim that AI safety is solved draws fire from former insiders(4 posts)→
More from AGI Musings
- New working paper: AI can forecast treatment effects in economics from pre-registration documents — soumitrashukla9 · 2026-10-29
- 'Taste' Is Mostly the Ability to Tell Reality From Marketing Materials — generativist · 2026-10-29
- Advertiser who spent $500k/mo on Google Ads says Search growth is artificial amid LLM cannibalization — SumitGup · 2026-10-29
- Billionaire foundations, not corporations, are the better misalignment example — EigenGender · 2026-10-29
- Mark Cuban: the defining divide is people who use AI to learn vs. to avoid thinking — MazMansoor · 2026-10-29
- Everyone talks about AI but they mean completely different things — almmaasoglu · 2026-10-29