Lessons learned on language model safety and misuse

OpenAI · Thu, 03 Mar 2022 08:00:00 GMT

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