Leaked Fable 5.1 system prompt compared against ChatGPT in behavioral test
doodlestein · x · 2026-09-02
A post compares the output of ChatGPT against a prompt excerpt derived from the recently leaked Fable 5.1 system prompt. The comparison highlights differences in response styles and tendencies, sparking discussion about model alignment and jailbreaking behaviors. Such tests are often used to evaluate variations in safety guardrails and instruction following across different models.
Related event: Leaked Fable 5.1 System Prompt Sparks Claude vs ChatGPT Comparisons(2 posts)→
More from Models
- Claude Fable 5.1 Replicates and Extends Research; Mythos Writes Custom Kernels — BenBlaiszik · 2026-09-02
- OpenAI's Astra Cybersecurity Model Reaches 'Critical' Threshold — alexcovo_eth · 2026-09-02
- Developer finds flaws in CritPt benchmark scores — scaling01 · 2026-09-02
- MiniMax-M3 hits 2,274 tok/s in Cacheon arena, focused on enterprise end-to-end testing — const_reborn · 2026-09-02
- Claude flags garden keep-out zones, suggests reorienting camera for safety — _Stocko_ · 2026-09-02
- WSJ: Google's New 3.8 Flash Model Narrows Coding Gap with Opus 5 — Charuru · 2026-09-02