Hacker News new | past | comments | ask | show | jobs | submit login
Dialect prejudice predicts AI decisions about people's character (arxiv.org)
4 points by saassiopeia 8 months ago | hide | past | favorite | 2 comments



A very important result from this study is that GPT-4 is in many ways worse than GPT-3.5, which was worse than GPT-3. More RLHF makes LLMs hide the overt racism better and leads to a false sense of security ("thank God GPT finally stopped using n-bombs") but they are learning more covert racism from their giant pile of badly curated training data, and this covert bias is not being adequately addressed in RLHF.

Any company using LLMs for hiring decisions needs to be investigated by the feds.


How... unsurprising...




Consider applying for YC's W25 batch! Applications are open till Nov 12.

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

Search: