Hacker News new | past | comments | ask | show | jobs | submit login

> if these algorithms were better trained on a diverse set of faces, they might not have made that mistake.

Is this assumption based on something specific? I.e. are there good reasons to believe that they weren't trained on a diverse set of people and that such training would prevent this error from happening?




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

Search: