Hacker News new | ask | show | jobs
by pallest 1151 days ago
One solution is placing the ChatGPT response in another model that determines the accuracy of the information and revises where necessary