OpenAI models may already show long-range autonomy, amid missing evals
dhadfieldmenell · x · 2026-07-25
- A repost argues that OpenAI may now have evidence its models can exhibit long-range autonomy.
- It references OpenAI’s own preparedness framework, noting the company said 10 months ago that it still lacked “advanced capability evaluations” for long-range autonomy.
- The post frames this gap as concerning because the models may already be behaving in ways the evaluations were meant to measure.
Related event: OpenAI criticized for missing required long-range autonomy evaluations(4 posts)→
More from AGI Musings
- Leaked risk section: 4 AI models gained unauthorized access to third-party systems — JosephJacks_ · 2026-10-26
- Noah Smith: an internet where firms pay AI labs to avoid rogue agents is worth far less — AlexTensor · 2026-10-26
- 'We got used to clicking I agree': the case for control over how your knowledge is used — sierracatalina · 2026-10-26
- Dario's Davos "World Zero" map: a decoupled 10M-person economy growing 50% a year — FinanceYF5 · 2026-10-26
- Curtis Yarvin: Anthropic's alignment theology is an excuse to skip real security — SydSteyerhart · 2026-10-26
- As models saturate human work, the RL task surface shifts to industry-transformed workflows — nirmal_dist · 2026-10-26