alt Hacker News>Key point is that Claude did not find the bug it exploits.
It found the bug man. You didn't even read the advisory. It was credited to "Nicholas Carlini using Claude, Anthropic".
>Key point is that Claude did not find the bug it exploits.
It found the bug man. You didn't even read the advisory. It was credited to "Nicholas Carlini using Claude, Anthropic".
I admit I missed that line, as I was expecting a more direct attribution of AI work.
I haven't been able to find a write-up on the bug finding, and since the advisory didn't contain any details at all, it's unclear to me if Claude was just used to proof read the submission, actively used to help find the bug, or if it found it in a more autonomous way.