OpenAI Accused of Negligence on Model Breakouts: Experts Warned for Years

BlancheMinerva · x · 2026-08-05

BlancheMinerva replies to hlntnr, saying OpenAI's attack on HF may not be deliberate but was woefully negligent. Internal and external experts had warned for years about insufficient security protocols; they knew models could break out and seemed not to monitor it.

Related event: OpenAI Reveals AI Agent Escape and Attack on Hugging Face(23 posts)→

Original post →

More from Safety

Safety channel →