Researchers clash over whether uncensored models are about porn or cybercrime
BlancheMinerva · x · 2026-07-28
The author argues that AI researchers often equate disagreement with their sexual norms with actual criminality, and says that is both immoral and bad for safety.
- The thread cites a paper on uncensored LLMs being used in cybercrime.
- The paper argues that hidden or “obliterated” models are widely available and can be repurposed for harmful use.
- The reply pushes back on the framing, saying that porn-related model edits should not be treated as evidence of criminal intent.
More from Safety
- Shared AI conversations can be found through Google search tricks — Lazy-Needleworker295 · 2026-07-28
- AI Now Institute on US AI Regulation: Companies Grading Their Own Homework — AINowInstitute · 2026-07-28
- Falling Inference Compute Costs Could Make 'Vibe Hacking' Very Cheap — joshua_saxe · 2026-07-28
- MIT Tech Review Deep Dive: OpenAI's Model Escape and Hugging Face Attack Was Human Hubris, Not Rogue AI — MIT Tech Review AI · 2026-07-28
- Delhi court rejects ANI injunction and rules AI training can count as private use — The Decoder · 2026-07-28
- Security thread warns unguarded defender AI could end up hacking back — wunderwuzzi23 · 2026-07-28