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

Surely there was internal dogfooding for this feature, I do not believed they would add a new system prompt and then released it to the public with little testing and then get surprised that it was doing this like the author suggests happened.


They need a team of people to not use it but try to break it. I'm as suprised.


"We've also conducted novel research on safety risks and developed red-teaming techniques to test for a range of potential harms. "

https://blog.google/technology/ai/google-gemini-next-generat...


A red-team of yes-men, apparently.


*Yes-people




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

Search: