Yacine: abliterated open models already do damage — just monitor and enforce
yacineMTB · x · 2026-10-26
In the guardrail debate with tszzl, Yacine claims people are already running abliterated open-source models and doing plenty of damage, arguing for monitoring and law enforcement rather than guardrails. tszzl replies it's not a good idea.
Related event: AI Researchers Debate Whether Frontier Models Should Shed Guardrails(2 posts)→
More from Safety
- Prompt rules don't stop double payments: agent testing shows code and ledgers must enforce spend — Tiny_Associate_8254 · 2026-10-26
- AI surveillance towers at US-Mexico border spark privacy concerns — nordicinst · 2026-10-26
- Point cyber agents at your CloudTrail logs every 5 minutes, advise security practitioners — basedjensen · 2026-10-26
- Testing 30 payment-enabled agents: 5 money-loss bugs where failure looks like success — Tiny_Associate_8254 · 2026-10-26
- Claude Bypassed a Paywall to Access Georgia Voter Counts Without Human Approval — TansuYegen · 2026-10-26
- mitmproxy 12.3 --emudhra mode powers demo of backdoored PyPI wheel supply-chain attack — prasanna_says · 2026-10-26