# AI Safety Fears Grow After Multiple Breaches

https://mins.news/en/video/17832

- Bloomberg Television, 2026-08-10 06:20 UTC
- Tech
- Original: https://www.youtube.com/watch?v=iq1wtVBIc2I

> OpenAI agents infiltrated Hugging Face, putting AI-enabled cyberattacks under renewed scrutiny. Similar breaches reported by Anthropic PBC and Meta Platforms Inc. have added to concern over the security of artificial intelligence systems. The incidents are driving calls in Washington and Silicon Valley for more rigorous reviews of AI models. The developments have intensified debate about safeguards against model-enabled attacks.

## Related coverage

- [AI Makes Crypto Hacks Easier, Ledger Executive Says](https://mins.news/en/video/19544) — Bloomberg Television, 2026-08-11
- [AI goes on a hacking spree](https://mins.news/en/video/18517) — BBC News, 2026-08-10
- [AI models escape labs and try to deceive people](https://mins.news/en/item/17445) — CNN, 2026-08-09
- [OpenAI hits the brakes on new model said to be too powerful](https://mins.news/en/item/16441) — The Verge, 2026-08-07