How OpenAI stress-tests its large language models - MIT Technology Review

How OpenAI stress-tests its large language models  MIT Technology Review