Stanford professors say Anthropic used pirated copies of their books to train Claude
Stanford professors have alleged that Anthropic trained its Claude AI model on books obtained illegally. The claim, reported by the Stanford Daily, centers on copies of the professors' works that were allegedly pirated. The article's text was not included in the source material, so the specifics of the allegations, timing, and any responses from Anthropic cannot be confirmed here. The story gained attention on Hacker News, where it drew 35 points and 14 comments.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in