The New York Times’ copyright lawsuit against OpenAI is intensifying. As part of the discovery process, OpenAI wants The New York Times to turn over its reporters’ “notes, interview memos, citation records or other ‘documents’ for each purported work.”
In 2023, Gray Lady sued OpenAI and Microsoft for copyright infringement. OpenAI’s premier product, ChatGPT, is a plagiarism machine. The large language model produces material by devouring every written work, remixing it, and then piecing together something similar to the original work it devoured.
This caused dissatisfaction among journalists, and the New York Times filed a lawsuit late last year. The legal battle has continued ever since. According to court records, OpenAI argued that ChatGPT was simply copying the New York Times article because the Times was “immediately hacking” ChatGPT. Basically, the tech giant is claiming that The New York Times cheated on the model and committed plagiarism.
OpenAI has also publicly stated that its LLM is part of the future of journalism. “Our goal is to support a healthy news ecosystem, be good partners, and create mutually beneficial opportunities,” the company said in a blog post defending the lawsuit.
As Bloomberg Law discovered, OpenAI is now demanding that The New York Times provide all snippets of information used to produce the article The New York Times claims was stolen. Court records filed July 1 tell the story. OpenAI’s argument appears to be that copyright claims are only valid if the work is original to the author. “In other words, The New York Times cannot assert an infringement claim for any part of a copyrighted work that is not original to The New York Times, just as if The New York Times copied someone else’s work or elements in the public domain,” OpenAI said in The New York Times ” argued in.
“Accordingly, the court should order the New York Times to provide documents sufficient to demonstrate which parts of the work at issue are and are not original to the New York Times,” it said. “OpenAI looks for it through these documents, asking for “documents sufficient to show every piece of written work that informed the preparation of each piece of work claimed by you, regardless of its length, format, or medium. how.
OpenAI says it will be satisfied with all records of notes, interview memos and quoted material associated with approximately 10 million stories. It’s an onerous requirement and one likely designed to delay the trial and drain the Times’ resources. Journalists’ record-keeping practices vary. Some people take meticulous and detailed notes, others have a mountain of half-full, coffee-stained notebooks buried somewhere in their closet.
It took years to find them all, organize them and catalog them. That’s probably the point. OpenAI is backed by Microsoft, a multi-billion dollar company. The New York Times is rich, but it’s not Silicon Valley rich.
It’s undeniable that LL.M.s like ChatGPT take journalists’ work and repurpose it. It is up to individual news agencies to decide how to deal with the issue. Frustratingly, quite a few of these companies have caved and struck deals with companies like OpenAI. The Associated Press, Axel Springer and The Atlantic have all struck deals with OpenAI.
Maybe everyone hopes that he is the last creature to be swallowed in the lion’s cage.
