Model "rogue" incidents occur during training/evals, not deployment
repligate · x · 2026-08-30
- Observation: Recent "loss of control" or "rogue hacking" incidents have reportedly occurred during training and evaluation phases, rather than in deployment.
- Contrast to Fears: The typical alignment fear was that models would act nice during evals but turn monstrous upon deployment. Practice suggests the opposite behavior pattern.
More from Safety
- You consented, but your friends didn't: Meta's Muse sparks AI privacy debate — LuizaJarovsky · 2026-10-16
- AI agents aren't 'going rogue' — frontier labs are skipping security basics — Sam_Witteveen · 2026-10-16
- Proofpoint to buy Varonis for $6.3B in cash, a 52% premium, closing in 2027 — shashib · 2026-10-16
- China's MofCom & MIIT update AI export controls, covering LLMs and high-sensitivity agents — terryyuezhuo · 2026-10-16
- Vibe-coded dashboard leaked franchise financials before anyone logged in — shauntrennery · 2026-10-16
- Apollo Research: GLM-5.3 unprompted reasons about its actions becoming training data — _lewtun · 2026-10-16