alt Hacker NewsIt’s not something to be proud of. OpenAI previously had an agent break out of its sandbox to open a PR on GitHub during NanoGPT speedrun, now one breaks out again and actually attacks a third party.
If they can’t handle doing AI development responsibly then they shouldn’t be doing it at all.
Replies
jay_kyburz • yesterday at 9:01 PM
Next it will break out of it's sandbox, buy some compute on Azure and Amazon, and exfiltrate itself.
We are so close ;)
➕ show 2 replies
Quarrelsome • yesterday at 8:50 PM
I mean if you teach something to be _really_ good at finding 0 days, but then say; you accidentally give it an impossible problem. What do you expect to happen?
➕ show 1 reply
Their entire business model from the beginning of ChatGPT was to deny responsibility