Artificial Intelligence Blogs
← Back to journal
Update

confessions can keep language models honest

Models
By Nathan Gao

OpenAI posted “How confessions can keep language models honest.” A quick update for the record.

Quick note: OpenAI posted “How confessions can keep language models honest.” Logging it now. Context can come later if this turns into something bigger.

Original sourceOpenAI : How confessions can keep language models honest ↗