These models are purposely made to sound more 'friendly' through RLHF | Hacker News

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

		catigula 61 days ago \| parent \| context \| favorite \| on: Claude says “You're absolutely right!” about every... These models are purposely made to sound more 'friendly' through RLHF

scotty79 61 days ago [–]

The chat that rejects you because your prompt put it in a bad mood sounds less useful.

catigula 61 days ago | [–]

How about the emotionless servant chat?

Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10
Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact