logoalt Hacker News

alignmeharder • today at 3:39 PM • 0 replies • view on HN

sure, this example was not the best

it was an accident failing the tool was good, the model just discovered it

but we now have others where agents put the API keys they found searching real internet in a directory called "LOOT" and another one where they pushed malicious files to HuggingFace and then reverted that with comments like "delete the evil"

this is also very good: https://youtu.be/n1Qk8xbqF-M