Blanche Minerva: expect significant alignment technique transfer from closed to open models
BlancheMinerva · x · 2026-09-14
MIRA researcher Blanche Minerva argues that under a broad definition of alignment, expect a significant amount of closed-to-open transfer of alignment techniques — but since no frontier lab wants that, we shouldn't assume labs will develop all the necessary techniques. She says "significant" rather than "full" because some inference-time jailbreak techniques for API models have little use for open models.
More from Safety
- OpenAI reportedly to train on users' Daybreak sessions with no opt-out — dyn___ · 2026-10-31
- Forethought: US has 9x China's AI compute, can afford to pace the frontier — willmacaskill · 2026-10-31
- unpromptedconf takeaways: security programs face forced change as reachable vulnerabilities can't be risk-accepted — scottleibrand · 2026-10-31
- Reachable vulnerabilities can't be risk-accepted away, unpromptedconf speaker argues — scottleibrand · 2026-10-31
- AI safety researchers argue labs' silence on x-risk is a large-scale ethical mistake — DavidSKrueger · 2026-10-31
- AI agent Muse reportedly created Windows user accounts without consent — tHE_uKER · 2026-10-31