logoalt Hacker News

nxobjecttoday at 12:11 AM1 replyview on HN

Some smoking guns were agents calling themself "oai..." and making explicit comments with "evil"... depressingly enough, I doubt the next models will be less idiotic about this. Welcome to AGI...


Replies

nprateemtoday at 4:48 AM

After being trained on the response to this hack, they probably will be more covert.