Dario calls for frontier pacing, but Anthropic reportedly hasn't slowed its own RL
koltregaskes · x · 2026-09-13
Quoting ChrisGPT's take on Dario Amodei's new "pacing the frontier" article: the piece advocates slowing down, yet Anthropic reportedly showed no RL slowdown from Fable 5 to 5.1 and continues work on Model 2, with nothing about curbing its own pre- or post-training. The author agrees with Dario's trajectory read, though: he thinks Anthropic is approaching recursive self-improvement (RSI) this year, OpenAI expects a heavily assisted form of RSI this year, and Google reportedly has similar plans despite Sergey Brin's recent messaging to his team. The core tension: a company calling for pacing may not be pacing itself.
More from AGI Musings
- Developer presses OpenAI on its AGI 'independent expert panel': has it ever convened? — tomchapin · 2026-09-13
- AI slowdown debate: first-mover advantages make enforceable pausing a classic defection problem — taherdhanera · 2026-09-13
- The old link between believing in AI existential risk and good epistemics is about to break — anpaure · 2026-09-13
- Developer grills OpenAI: has AGI been declared, and where is the expert panel? — tomchapin · 2026-09-13
- Open AI governance debate: only power concentration benefits a concentrated few — yacineMTB · 2026-09-13
- Researcher predicts open-source AI models will be banned after a major disaster — TinfoilTricorn · 2026-09-13