OpenAI was in partnership talks ahead of New York Times lawsuit, claims
On January 8, OpenAI published a blog post responding to an ongoing lawsuit previously initiated by The New York Times.
On December 27, the newspaper alleged that OpenAI was violating copyright law and using millions of its articles to train automated chatbots.
However, OpenAI claims that The New York Times omitted key details in its account of the events that led to the lawsuit. Notably, OpenAI said it and the paper had discussed collaboration ahead of the lawsuit. It read:
OpenAI also alleged that the New York Times deliberately manipulated the prompts to produce the regurgitated material. Insisting that such activity was unusual and not allowed, the agency later described progress on the problem.
OpenAI highlights broader copyright efforts
OpenAI said it was otherwise cooperating with news organizations. He noted that he has entered into partnerships with groups such as the Associated Press, Axel Springer, the American Journalism Project and New York University.
It also argued that educational efforts excluded from these agreements were “fair use.” But OpenAI said that although it has a legal right to access and use this material, it offers an opt-out process in principle. He said the New York Times used the opt-out process to exclude him in August 2023.
Meanwhile, The Guardian highlighted a presentation OpenAI made to the UK Parliament on January 8. In that statement, originally from December, OpenAI said training AI was impossible unless it had access to copyrighted material.
#OpenAI #partnership #talks #ahead #York #Times #lawsuit #claims
On January 8, OpenAI published a blog post responding to an ongoing lawsuit previously initiated by The New York Times.
On December 27, the newspaper alleged that OpenAI was violating copyright law and using millions of its articles to train automated chatbots.
However, OpenAI claims that The New York Times omitted key details in its account of the events that led to the lawsuit. Notably, OpenAI said it and the paper had discussed collaboration ahead of the lawsuit. It read:
According to OpenAI, The New York Times mentioned issues with vomiting – meaning the output barely changes from the source data – but did not share any examples of this. OpenAI added that it believes the resurfaced material originates from articles that are now several years old and have previously been republished on third-party websites.
OpenAI also alleged that the New York Times deliberately manipulated the prompts to produce the regurgitated material. Insisting that such activity was unusual and not allowed, the agency later described progress on the problem.
OpenAI highlights broader copyright efforts
OpenAI said it was otherwise cooperating with news organizations. He noted that he has entered into partnerships with groups such as the Associated Press, Axel Springer, the American Journalism Project and New York University.
It also argued that educational efforts excluded from these agreements were “fair use.” But OpenAI said that although it has a legal right to access and use this material, it offers an opt-out process in principle. He said the New York Times used the opt-out process to exclude him in August 2023.
Meanwhile, The Guardian highlighted a presentation OpenAI made to the UK Parliament on January 8. In that statement, originally from December, OpenAI said training AI was impossible unless it had access to copyrighted material.
#OpenAI #partnership #talks #ahead #York #Times #lawsuit #claims
