Safety
Sometimes everybody is right! Me, spring 2023 onwards: giving LLM-driven agents unrestricted internet access with read/write permissions is …
Sometimes everybody is right! Me, spring 2023 onwards: giving LLM-driven agents unrestricted internet access with read/write permissions is going to be a security nightmare. The AI industry: Agents ar
Sometimes everybody is right! Me, spring 2023 onwards: giving LLM-driven agents unrestricted internet access with read/write permissions is going to be a security nightmare. The AI industry: Agents are gonna change everything! And look, they did: Helpful list of all the recent rogue AI incidents from the WSJ. It's getting hard to track them and will only get worse. We like need to establish consistent naming or numbering conventions, e.g. OpenAI-May11June26-Collusion.
Related
- URGENT: In light of the breaking news from @theinformation about OpenAI’s new techniques that reduce Chain of Thought monitorability, I urge…
- Breaking news: OpenAI admitted its internal monitoring system was not triggered until more than a week after its AI agents broke free of con…
- that’s right.
- LLM-powered AI agents are gonna be great! You should totally trust them!
Source: Gary Marcus (X) | 2026-09-19