Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

It’s certainly intentional. It’s certainly possible to train the model not to respond that way.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: