OpenAI Accuses New York Times of ‘Hacking’ ChatGPT

OpenAI has petitioned the court to dismiss part of the New York Times (NYT) copyright lawsuit, alleging the newspaper “hacked” ChatGPT to fabricate evidence. This was reported by Reuters.

“What OpenAI calls ‘hacking’ is merely using its own products to find evidence that they stole and reproduced NYT’s copyrighted work,” responded the newspaper’s lawyer, Ian Crosby.

In December 2023, the publication sued OpenAI and its largest investor, Microsoft, accusing them of using millions of its articles to train chatbots without permission.

The New York Times complaint cites several instances where ChatGPT provided users with nearly complete copies of articles. The publication accuses the tech companies of attempting to create a substitute for newspapers.

“NYT cannot prohibit AI models from acquiring knowledge of facts, just as another news organization cannot prevent a newspaper from reprinting stories in which it did not participate,” responded the creators of ChatGPT.

OpenAI representatives added that it is impossible to obtain verbatim texts from articles through normal use of the chatbot.

Currently, it remains unresolved by law whether AI training and the works created from it constitute copyright infringement.

Earlier in February, users complained about nonsensical responses from ChatGPT.