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

This is anecdotal, but the article used ChatGPT to score the sentiment. I’ve noticed that ChatGPT tends to “hallucinate” positive sentiment where there is sufficient nuance but a person would interpret it as overall negative[^1]. I however haven’t tested that bias against more brazen statements.


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

Search: