Critics call Anthropic two-faced: warns against recursive self-improvement while touting it as edge
wfithian · x · 2026-11-20
Peter Wildeford highlights a contradiction in Anthropic's statements: it once warned that recursive self-improvement "could outrun our ability to understand and control these systems, and so must be pursued very carefully, if at all" — yet also declared "We're excited to do recursive self-improvement. It is our key competitive advantage!"
Daniel Tan, whose post was retweeted here, goes further: Anthropic is fundamentally two-faced — one part earnestly focused on safety, another straightforwardly power-seeking, with the latter apparently controlling what the company actually does.
More from AGI Musings
- Stanford Digital Economy Lab launches monthly AI Economic Indicators brief tracking jobs and adoption — erikbryn · 2026-11-20
- Blogger predicts GTA 6 will serve as the simulation environment for 2027's AI agents — rachittshah · 2026-11-20
- Post claims Claude Opus 5.5 found 872 major errors in published astrophysics papers, drawing pushback — fredahshi · 2026-11-20
- OpenAI Foundation board member Paul Christiano warns AI takeover could unfold in months — EvanHub · 2026-11-20
- Devil's Advocate: Should AI Resisters Also Forfeit AI's Upside? — Afinetheorem · 2026-11-20
- Why Google moves slower on AI: lawsuits and product survival as natural pacing — shakoistsLog · 2026-11-20