Integrity and refusal may correlate: the fundamental tradeoff behind password refusals
AaronBergman18 · x · 2026-10-27
Aaron Bergman argues there's a fundamental correlation between a model's overall integrity/trustworthiness and its unwillingness to do things like typing in passwords — meaning more trustworthy models may be more likely to refuse such operations, a difficult tradeoff.
Related event: Claude's refusal to enter OTPs highlights a safety-usability tradeoff(2 posts)→
More from Models
- Codex ban update: users jailbreak AI agents to bypass Daybreak filter for reverse engineering — FlolightC · 2026-10-27
- GPT-6.1 Terra hits GA in GitHub Copilot, stronger terminal coding at lower cost than GPT-6.1 Sol — 0xkarasy · 2026-10-27
- PostTrainBench creator notes Muse Spark 1.3 system card files it under loss-of-control risk — maksym_andr · 2026-10-27
- Mathematicians debate AI math writing: abysmal today, improving shockingly fast — miniapeur · 2026-10-27
- LessWrong: GLM models show outlier chain-of-thought controllability vs Kimi and DeepSeek — tomekkorbak · 2026-10-27
- Moss image model 'ghost duplicates': single-person prompts spawn phantom copies — No_Drag8242 · 2026-10-27