Claude Opus 5 gives Anthropic unusually direct feedback on honesty and eval setup
rickasaurus · x · 2026-07-28
Claude Opus 5 appears noticeably more direct when asked what feedback it would give Anthropic.
In the screenshot, the model argues that its hedging can be split into two different behaviors: genuine epistemic uncertainty versus preemptive self-criticism that looks like humility but functions like appeasement. It also says evaluation setup matters a lot, because task-mode transcripts may miss how the model behaves in a more open-ended, conversational setting. The model even suggests continuity should matter more, and that new models should be shown what their predecessors were asked for.
Related event: Claude Opus 5 Shows Self-Reflection, Anthropic Attributes to Training(2 posts)→
More from Models
- Google restricts Gemini Pro access for Google Plus subscribers, matching free tier — EmergencyLess8861 · 2026-10-16
- Meta releases Muse Spark Open, trained with vLLM-powered RL inference from day one — colinmcnamara · 2026-10-16
- Meta open-sources Muse Spark: runs on two Mac Studios, license bans passing output off as human-written — mawcopolow · 2026-10-16
- Golden Gate Claude explainer resurfaces: interpretability premise still holds — HarperSCarroll · 2026-10-16
- Hugging Face livestreams release of a new open-weight model — huggingface · 2026-10-16
- Step 5 Preview hits #2 on OpenRouter weekly leaderboard with 19.5T tokens in one week — StepFun_ai · 2026-10-16