AI safety papers keep conflating sexual content with actual criminal abuse
BlancheMinerva · x · 2026-07-28
The poster argues that many papers conflate sexual impropriety with criminal activity, and says the quoted paper does exactly that at length.
The reply it responds to draws a line between two different concerns: people are not mainly worried about models generating smut, but about models automating cybercrime, scams, and similar abuse.
The core issue here is how AI safety research frames harmful content and whether it muddies very different risk categories.
Related event: Study on 229 Uncensored LLMs Sparks Debate on Safety Boundaries(7 posts)→
More from Safety
- Meta Muse's first suggested name matches user's childhood dog, raising privacy questions — matt_slotnick · 2026-09-23
- Open-source advocates call doom narratives a regulatory moat against open weights — AlexTensor · 2026-09-23
- AI safety will follow engineering tradition: formal proofs for simple cases, evals for complex — burny_tech · 2026-09-23
- Stochastic Parrots authors rebut AI-pause letter: focus on present harms, not sci-fi risk — marigo · 2026-09-23
- Devs mock labs' cyber-enabled Claude/GPT testing as 'felonies sold as safety research' — ctjlewis · 2026-09-23
- Okta launches Human Principal, binding AI agents to verified humans via World ID — BecauseCulture · 2026-09-23