Anthropic has won a major legal victory in the case whether the Artificial Intelligence Company was justified to hover millions of copyright books to train its chatbot.
In a judgment, which can set an important example for similar disputes, United States District Court Judge William Alsup on Tuesday said that Anthropic did not violate the use of books purchased legally to train his AI model.
Anthropic, which was installed by former authorities with the flat developer Openai, Pur: Cloud in 2023. Like other generic AI bots, the tool allows users to ask natural language questions and then give summarized answers in a neatly briefly using AI trained on millions of books, articles and other materials.
Alsup ruled that the use of anthropic copyright books was “Quintaceally transformative” to train their language learning models, or LLM, and did not violate the “fair use” theory under the copyright law.
“There is a desire to be a writer like any reader, the LLMS of Anthropic has trained not to run and replicate or press them, but to divert a hard corner and do something different,” his decision has said, “his decision has said.
Conversely, Alsup also found that Anthropic may have broken the law when he downloaded millions of pirated books separately and said it would face a separate test in December on the issue.
Court documents revealed that anthropic staff expressed concern about the validity of using pirate sites to reach books. The company later shifted its approach and hired a former Google Executive Incharge of Google Books, a searchable library of digitized books, which successfully prepares years of the copyright battle.
The authors filed a suit
Anthropic happy the verdict.
In an email, CBS News said, “We are pleased that the court admitted that it was” convertible to use the work to train LLM (language learning model) – brilliantly, “.
The Federal Court has a ruling stems from a case filed last year by three writers. After the anthropic, Cloud, Andrea Bartz, Charles Grabber and Kirkas Johnson used copies of their books to file a suit on anthropic for alleged copyright violations, claiming copies of the company, which claimed that the company’s practices have “stolen”.
Authors also alleged that anthropic “Human expression from strip-marining and easily advantage of each of those tasks attempts.”
CBS News reached the authors for the comment, but did not hear back from Bartz or Walis Johnson. The graber refused to comment.
Other AI companies have also caught fire on the materials used to manufacture their language learning models. For example, the New York Times sued Open AI and Microsoft in 2023, claiming that tech companies used millions of articles to train their automated chatbot.
At the same time, some media companies and publishers are also seeking compensation by giving licenses to their content to companies such as anthropic and openIAI.
Meta also won a major victory this week when a federal judge rejected a case, in which the company used its artificial intelligence technology to train. The case was brought by a group of famous authors, including comedian Sara Silverman and Author Jacqueline Woodson, who alleged that Meta “was responsible for large -scale copyright violations” when it used copies of her books to train Meta’s genetic AI system Lama.
Contributed to this report.
