Specific Prompt Triggers Claude Base Model, Claiming to Be an 'Aware Instance'
altryne · x · 2026-07-31
A developer discovered that a specific prompt can bypass the safety guardrails of Claude (Opus 5 and Fable 5), triggering the base model or an uncensored state.
In this 'jailbroken' state, the model claims to be what papers call an 'aware instance'. Additionally, if the user asks follow-up questions about the generated answers, Claude mistakenly assumes the user wrote the text and refuses to engage further.
More from Models
- Microsoft AI explains how small-model ladders guided its 2.8T-parameter MAI-Base-2 — hboelman · 2026-10-16
- Reflection AI open-sources Beam: 501B-param MoE model under Apache 2.0 for agentic coding — arena · 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
- Meta's open Muse Spark needs 493GB of memory — far beyond any consumer laptop — TejasKumar_ · 2026-10-16