§ feed · storyline

Evaluating fairness in ChatGPT

OpenAI publishes analysis of how ChatGPT responses vary based on user names, using AI research assistants to examine fairness in model outputs.

Oct 15 · 12:00:00 · primary fetch1 sourceupdated Oct 15 · 12:00:00

We've analyzed how ChatGPT responds to users based on their name, using AI research assistants to protect privacy.

§ sources1 publication · timeline below