Lessons learned on language model safety and misuse 文章

OpenAI Blog2022-03-03BLOGen

摘要

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