OpenAI publicly responded to the copyright lawsuit he filed The New York Times last month. And she didn’t hold anything back. The creators of ChatGPT They assure that the complaint has no merit and accuse the American media of manipulate your chatbot to show that he had allegedly copied his articles.
OpenAI said that The New York Times “It doesn’t tell the whole story.” “It appears that they intentionally manipulated the prompts, which often included extensive excerpts from articles, so that our model [los] regurgitate,” the company says in a release.
The New York Times is suing both OpenAI and Microsoft, a key investor in this startup and which uses its technology in several products. The media, one of the most important in the world, accuses the two companies of training ChatGPT’s artificial intelligence by copying its journalistic texts without authorization.
OpenAI now says that they had requested The New York Times to send them some examples of plagiarized content. “But they repeatedly refused to do so, despite our commitment to investigate and resolve any problems,” they say in the text.
According to Sam Altman’s company, the New York media allegedly manipulated ChatGPT into “regurgitating” specific fragments of his articles. Something I would have achieved through prompts long and extremely specific text descriptions. «Interestingly, the regurgitation that induced The New York Times They appear to come from years-old articles that have proliferated on multiple third-party websites,” OpenAI says on its blog.
OpenAI, surprised by the complaint The New York Times
OpenAI claims it held negotiations with The New York Times until last December, with the aim of paying for authorization for the use of its content. «His demand of December 27, which we learned about by reading the NYT“It was a surprise and a disappointment for us,” says the Californian firm.
The company admitted that A few months ago it removed a ChatGPT feature, called Explore, that played—supposedly unintentionally—protected content. Also, he explained to the American media that, like any single source, “its content did not contribute significantly to the training of our existing models and would not have sufficient impact for future training.”
But accusations of plagiarism do not come only from The New York Times. OpenAI and Microsoft were also sued by a group of prominent authors, which includes several Pulitzer Prize winners. Famous writers such as John Grisham, George RR Martin —creator of the original books of Game of Thrones—and Jonathan Franzen.
OpenAI has also recognized that models like ChatGPT cannot be created without the use of copyrighted material. “It would be impossible,” he said in a presentation to the upper house of the British parliament. “Copyright today covers virtually every type of human expression,” he acknowledged in a document dated last December.
Sam Altman’s team wants to continue negotiating
Despite everything, OpenAI says that wants to continue negotiating with the American media. “Still, we are hopeful for a constructive partnership with The New York Times and we respect its long history”, assures the company.
OpenAI has already begun to formalize content usage agreements with others. Recently managed to sign an alliance with American Journalism Project and the German publisher Axel Springer. And in July of last year, he received permission to use the news archive of Associated Press.
“We wait continue collaborating with news organizationshelping to elevate its ability to produce quality journalism by taking advantage of the transformative potential of artificial intelligence,” the company highlighted.