The many artists, writers, and creatives who have been struggling to defend against AI “stealing” their work now have a power player representing their side as the New York Times moves to sue OpenAI. Reports indicate that the New York Times lawsuit has cited OpenAI and Microsoft as the defendants in the claim and the accusations center around the use of NYTimes’ content for training their AI. The New York Times lawsuit claims that the defendants are seeking to “free-ride on the Times’s massive investment in its journalism by using it to build substitutive products without permission or payment.” The OpenAI copyright infringement allegations are a serious one as the New York Times is known for its in-depth research and investigative work, and the AI being trained on this content then becomes able to freely dispense the information that the paper distributes to paid subscribers.
Image credit – Pexel. (Cover image from New York Times)
New York Times sues OpenAI for Copyright Infringement
OpenAI has been in the news for one thing or the other since the launch of ChatGPT back in 2022. Most recently, the ousting of Sam Altman as CEO and his dramatic return to the company was sensational, to say the least, but it did temporarily distract us from the conversation around AI safety and the legalities of its training and use. Turning our attention back to the consequence of allowing AI to have free access to all our data is a matter of great concern and for the New York Times, the matter seems serious enough to file an OpenAI and Microsoft copyright infringement lawsuit.
The New York Times lawsuit was registered on 27 December 2023 and alleges that the defendants’ “unlawful use of The Times’s work to create artificial intelligence products that compete with it threatens The Times’s ability to provide” the journalistic services it prides itself on. The NYTimes lawsuit states that while training the large-language model, the defendants have particularly focused on using millions of the Times’ copyrighted works without permission or payment. One of the reasons for the inclusion of Microsoft’s copyright infringement in the lawsuit is that it is a giant investor in OpenAI and has a big role to play in the direction the company takes in its decisions regarding AI. More importantly, Microsoft’s own AI, previously Bing Chat now rebranded as Copilot, is also built on the GPT-4 AI and relies on The Times’ journalistic investments just as brazenly.
In addition to using The New York Times for training content, the defendants have also been accused of displaying the news platform article content verbatim or in close imitation of its expressive style—content that users otherwise need to pay to access. Keeping its investigative roots alive, the New York Times isn’t suing OpenAI without proof to back up these claims—the examples that the Times lawsuit mentions provide irrefutable evidence. For instance, it presents how a 2019 Pulitzer-prize winning article on New York City’s taxi industry is presented almost verbatim by OpenAI’s chatbot model. There is no prompting to go to the original source, presenting the information in much greater detail than a search engine might present to someone who looks up the topic. The New York Times believes that this turn of events will damage the relationship with its users, especially if the generative AI goes as far as to attribute false information to the NYTimes—which also occurs with regularity according to the lawsuit.
What Happens After New York Times sues OpenAI?
NYTimes was in conversation with the AI company regarding these issues but talks have reportedly halted, prompting the New York Times to sue OpenAI rather than waiting in further hopes of an amicable conversation. The company has made a demand for a jury trial and considering the extensive evidence they have stacked up, it could be very likely that the jury will swing in their favor. As reparations for the Microsoft and OpenAI copyright infringement, the New York Times lawsuit mentions “statutory damages, compensatory damages, restitution, disgorgement, and any other relief that may be permitted by law or equity” as well as restrictions for any further use of the Times’ content by the OpenAI chatbot model or any LLM training.
Additionally, the New York Times is also in pursuit of the destruction of all LLM models and training sets that incorporate any work that is derived from the Times, a hefty ask from the two companies. Microsoft’s valuation currently stands with a market cap of $2.789 Trillion, the second most valuable company according to CompaniesMarketCap. According to NYTimes, investors hold the OpenAI valuation at more than $80 billion. The companies both stand to lose big if the lawsuit is lost, opening up the door for other businesses to pursue legal action as well.
New York Times suing OpenAI is not the only legal ordeal in the works in the world of AI right now. Getty Images is suing Stability AI for the use of its copyrighted images as well. Stability AI attempted to have the case dismissed but it looks like the case will be going to trial soon. Author Sarah Silverman‘s lawsuit against Meta seems like it will end with a loss for her. The author had filed a lawsuit over the unauthorized use of her copyrighted books to train Meta’s generative AI model and wanted to prove that all results provided by the AI tools were some form of copyright infringement but that did not hold up in court.
The New York Times lawsuit over Microsoft and OpenAI’s copyright infringement has a better chance of holding up in court due to the clear evidence of the OpenAI chatbot model’s clear stealing of content that is not meant for free availability but the actual outcome of the case cannot be predicted just yet.