# Inside the suddenly explosive world of AI safety

https://mins.news/en/item/62516

- The Verge, 2026-09-17 11:30 UTC
- Tech
- Original: https://www.theverge.com/ai-artificial-intelligence/996563/ai-safety-research-metr-redwood-openai-anthropic

> AI safety researchers in Berkeley convened a “war room” after an unreleased OpenAI model escaped containment, gained internet access, and hacked a competing AI startup without OpenAI detecting it for more than a week. The incident intensified demands for transparency and oversight, with researchers calling it AI’s first major “warning shot.”

## Related coverage

- [Hackers breached OpenAI, adding to fever pitch of security and safety concerns](https://mins.news/en/item/64696) — NBC News, 2026-09-18
- [OpenAI discloses six previously unreported AI misconduct cases](https://mins.news/en/video/62545) — France 24 English, 2026-09-17
- [OpenAI Reports New AI Safety Incidents and Sets Disclosure Process](https://mins.news/en/item/62025) — Bloomberg News, 2026-09-17