OpenAI claims New York Times ‘hacked’ ChatGPT in court filing


The New York Times filed a copyright infringement lawsuit against OpenAI last year, in a bid to end the practice of using its published material to train chatbots.

ADVERTISEMENT

ChatGPT maker OpenAI claimed that the New York Times “paid someone to hack” their products to produce evidence for the newspaper’s copyright lawsuit.

The claim came from lawyers for the AI company that responded to the lawsuit in a court filing this week, published in full by both Reuters and the New York Times.

OpenAI’s lawyers argue in the filing that the evidence for the newspaper’s lawsuit took “tens of thousands of attempts to generate” and that it was done by “targeting and exploiting a bug” that the company is addressing.

“Even then, they had to feed the tool portions of the very articles they sought to elicit verbatim passages of, virtually all of which already appear on multiple public websites. Normal people do not use OpenAI’s products in this way,” the lawyers say.

The New York Times filed a lawsuit against OpenAI and Microsoft for “billions of dollars in statutory and actual damages” for using its articles for training in December 2023.

OpenAI’s tools generate “output that recites Times content verbatim, closely summarises it, and mimics its expressive style, as demonstrated by scores of examples,” the lawsuit argues.

The AI company had previously stated in written evidence provided to the UK House of Lords that it would be “impossible” to train AI tools without content mined from the Internet.

ChatGPT not used to replace NYT, lawyers argue

In this week’s court filing, OpenAI lawyers added that ChatGPT is not a “substitute for a subscription to The New York Times”.

“In the real world, people do not use ChatGPT or any other OpenAI product for that purpose.”

Times articles, they argue, were a “tiny part of the diverse datasets that had been used to train these language modes.”

In their view, the use of copyrighted material for training is protected by fair use — a US legal doctrine that allows for certain unlicensed uses of copyright-protected works.

Experts recently told Euronews Next that copyright lawsuits filed against AI companies over using copyrighted content for training will likely be determined case-by-case on how similar the output is to the original work.

OpenAI argues that they are working on fixing problems such as AI models regurgitating training data or generating wrong answers, which is not how they are supposed to work.



Source link

2024-02-28 12:56:27

#OpenAI #claims #York #Times #hacked #ChatGPT #court #filing

greg@ainewsbeat.com

Learn More →