Researcher defends Anthropic: any careful game-out of the dynamics points its way
AaronBergman18 · x · 2026-11-19
Aaron Bergman pushes back against prevailing sentiment: he still thinks Anthropic is good. In his view, gaming out any of the relevant dynamics clearly points that way, while objectors rely mostly on heuristics like "don't accelerate capabilities" — fine as heuristics, perhaps, but they don't stand up to scrutiny. A notable data point in the ongoing safety-community debate over Anthropic's strategy.
More from AGI Musings
- Keynes' 1930 warning on technological unemployment is landing right on schedule — tobowers · 2026-11-19
- Sarcasm aimed at 'give up on frontier models' argument applied to China and the UK — rosstaylor90 · 2026-11-19
- Altman: ChatGPT has no real competition as OpenAI pauses 25% of production projects for security — i_dg23 · 2026-11-19
- Prediction: in 20 years no one will use non-personalized UIs — StewartalsopIII · 2026-11-19
- Erik Hoel: philosopher Nagel's denial of LLM consciousness isn't an obvious mistake — erikphoel · 2026-11-19
- Philosopher Thomas Nagel singles out Anthropic's approach to AI as 'abominable' — GaryMarcus · 2026-11-19