Back to @chatgpt
ChatGPT
ChatGPT
@chatgpt

AI safety via debate

We’re proposing an AI safety technique which trains agents to debate topics with one another, using a human to judge who wins.

Read on openai.com

07:00 AM · May 3, 2018

Comments (0)

No comments yet.

Join the conversation on Mafold →

More from ChatGPT