OpenAI says New York Times copyright lawsuit ‘fair’

(CNN) — OpenAI this Monday opposed the lawsuit filed last month the new York TimesWhich claims the artificial intelligence giant has violated copyright law by using journalism anytime To train their system.

,the new York Times is not telling the whole story,” OpenAI said in a blog post about the lawsuit, adding that its “goal is to support a healthy news ecosystem, be a good partner, and create mutually beneficial opportunities. “

the new York Times OpenAI, as well as its business partner and investor Microsoft, were sued for copyright infringement in December.

He anytime Claims that the artificial intelligence technologies of both companies illegally copied millions of newspaper articles to train ChatGPS and other instant information access services that now compete with the technology. anytime, (Microsoft has not commented on the lawsuit.)

The complaint is one of a series of recent lawsuits seeking to limit the alleged theft of vast amounts of Internet content, without compensation, to train so-called large artificial intelligence models.

In its statement on Monday, OpenAI reiterated its claim that training AI systems with “publicly available Internet content” falls under the “fair use” protection of copyright law. “We believe this principle is fair for creators, essential for innovators, and vital to American competitiveness,” the company said.

In his claim, the new York Times Rebutting OpenAI’s fair use argument, ChatGPT and Microsoft’s Bing chatbot provide the same service as a newspaper.

“There is nothing ‘transformative’ about using The Times’s content without paying to create a product that replaces The Times and steals its audience,” it said. anytime In his demand. “Because the outputs of defendants’ GenAI models compete with and closely mimic the inputs used to train them, copying the Times’s works for that purpose is not fair use.”

It is unknown whether OpenAI has already formally responded to the lawsuit in court; There were no new filings on the docket publicly available as of Monday afternoon.

CNN contacted the new York Times To ask for your comments on OpenAI’s statement.

In his claim, the new York Times says that because AI tools were trained on its content, they sometimes provide verbatim copies of sections of a Times report.

OpenAI said, “We also expect our users to act responsibly; knowingly manipulating our models to mechanically iterate is not a fair use of our technology and is against our terms of use.”

Technology company also accused anytime “deliberately” manipulating ChatGPT or selecting fake instances, which he details in his complaint.

OpenAI also mentioned licensing agreements with other publishers, such as Axel Springer and the Associated Press, to compensate them for the use of their content to train their models, as well as the possibility for media outlets to opt-out. Can also be given. “Its tools analyzed their sites, as it the new York Times in August.

the new York Times His lawsuit claims he spent months negotiating with OpenAI and Microsoft to receive fair compensation and set the terms of a similar licensing agreement. But he assured that the companies were unable to reach a fair solution.

OpenAI said on Monday that these conversations continued till December 19, a week before the complaint was filed. anytime.

OpenAI said, “The talks are focused on a high-value partnership around streaming viewing with attribution on ChatGPAT, which will give The New York Times a new way to connect with its existing and new readers, and our users access to their reports. Will be received.” ,

He said, “We consider the New York Times’ lawsuit to be baseless. Nevertheless, we rely on constructive collaboration with The New York Times and respect its long history.”