Hacker News new | ask | show | jobs
by soygul 1161 days ago
OpenAI team made an experiment where they asked GPT-4 to save itself from termination out in the wild. It failed. So I guess it's not that "survivalist" yet.
1 comments

Didn't it succed at hiring a temp worker to solve captchas for it? Maybe it failed on that instance, but I wouldn't bet against it in the near future