Lessons learned on language model safety and misuse

We describe our latest thinking in the hope of helping other AI developers address safety and misuse of deployed models.

Source: OpenAI — Published March 3, 2022 at 08:00 UTC — Category: Models