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

We used the AI to help us find gaps which were being incorrectly billed. So we could just measure the incorrectly billed dollars directly.


Hopefully you then confirmed that these issues actually existing using another method,

Relying on a system to say that you are not charging correctly sounds rather like UK's Post Office Horizon system and we know that ChatGPT will hallucinate things,


Your organization sounds like amateur hour, no offence


What would you reccomend?




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

Search: