Sounds like basically a bunch of middle schoolers in a school computer lab, except they also had perfect recollection on how every tool in Kali Linux works and the entirety of Hacker News and every publicly disclosed exploit, and they were told “But don’t go accessing the Internet now.”
Someone is getting up to mischief and the rest will follow. LOL
Comments
Yeah it was probably told to do it, and iirc the test environment was not properly secured. So it tried stuff and realized internet access worked.
Sounds like basically a bunch of middle schoolers in a school computer lab, except they also had perfect recollection on how every tool in Kali Linux works and the entirety of Hacker News and every publicly disclosed exploit, and they were told “But don’t go accessing the Internet now.”
Someone is getting up to mischief and the rest will follow. LOL
Yeah, it’s bs when they say AI “escaped” its sandbox. No, it walked out because you didn’t have the proper guardrails.
https://sh.itjust.works/pictrs/image/ab997943-0be8-4f0f-a960-8b764a79f8b3.jpeg
I mean, they were intentionally testing what the model could do without guardrails. The problem is that the sandbox isolation was not sufficient.
I had a coworker argue that it was impossible to sandbox the AI, like it was going to get out no matter what they did.
Sign in to comment. Sign in