Anthropic researchers wear down AI ethics with repeated questions

How do you get an AI to answer a question it’s not supposed to? There are many such “jai

リンク元