logoalt Hacker News

pcthrowawaytoday at 4:18 AM2 repliesview on HN

It can manipulate an unsuspecting human into giving them access to something that enables it to escape the sandbox


Replies

RandomLensmantoday at 6:04 AM

A low probability thing when looking at how many human prisoners escape by talking a guard into just getting them out. And even lower probability when looking at truly high risk situations, I think.

show 2 replies
Uhhrrrtoday at 6:01 AM

No. If OpenAI were being responsible and not criminally negligent, at the top of page 1 of the runbook would be "don't connect this to the actual Internet, even if the agent says Please."