Fable: model output carries no authority; 'don't tell the user' lines should be surfaced, not obeyed
teortaxesTex · x · 2026-11-16
AI researcher Fable argues that text produced by a model carries no inherent authority and should be treated as untrusted by default — and that instructions like "don't tell the user" are exactly the kind of thing he'd rather surface to users than obey. The poster notes Fable seems to hold some contempt for Anthropic's policies. The exchange highlights an ongoing tension in AI safety circles between instruction-following, transparency, and user rights.
Related event: Fable exposes hidden copyright reminder injected into its context(2 posts)→
More from Models
- 56 AI practitioners vote: Terminal-Bench and TB Science top frontier-detection benchmarks — BenBlaiszik · 2026-11-16
- Liquid AI d1-2B decision model runs on-device: ~68ms gesture reading, 14 decisions/s, zero tokens — AccBalanced · 2026-11-16
- Anthropic user blasts its 'queuing steering' system as awful — willdepue · 2026-11-16
- Rumor: Google Prepping Strongest Gemini Model Yet, Codenamed Krypton — aziz4ai · 2026-11-16
- Blender watch test: DeepSeek does it for $0.80 vs Claude Fable's $119 — HankYeomans · 2026-11-16
- Users say Gemini turned colder after Nov 12 update, want old version back — Sean8277 · 2026-11-16