Getting your
Trinity Audioplayer ready... A senior executive at Microsoft believes A.I. systems built on the work of others will be seen as “an astonishing theft of unprecedented proportions” and possibly the greatest robbery of labor in human history, according to court documents unredacted Thursday in an ongoing Manhattan lawsuit.
The executive’s startling admission appeared in a newly unredacted filing in the landmark copyright infringement legal battle playing out in Manhattan Federal Court between newspapers such as the Daily News and The New York Times and tech companies Microsoft and OpenAI.
In a motion for summary judgment filed earlier this month, lawyers for the news outlets — who had pushed to uncover swaths of redactions shielding the tech companies — pointed to comments written by Microsoft’s Director of Applied Science, Brent Hecht, in a document soon after the suit was filed.
“[M]illions of people around the world will soon consider large models ‘hoovering up’ all their work to be an astonishing theft of unprecedented proportions,” which could pose the “largest theft of labor in human history,” Hecht was quoted as saying.
“[A]lmost no one intended for content they created to be used in this fashion, nor are they compensated for its use.”
Reached for comment, attorney Steven Lieberman, leading the Daily News’s defense, said Hecht’s comments prove the companies are knowingly committing theft, as opposed to operating within the parameters of the “fair use” doctrine as they claim.
“Throughout this case, defendants insisted that these documents be treated as confidential so that the public could not see them,” Lieberman said. “Well, now the cat is out of the bag. Finally, the world can see what OpenAI and Microsoft thought all along about the fairness of their own behavior.”
The newspapers’ motion lays out how Microsoft’s executives acknowledged the harm the company itself could suffer from providing A.I. chatbot users with the information they seek on the spot rather than redirecting them to the underlying source.
“Our AI content strategy has started a ‘doom loop’ that will hurt the performance of our models and the entire web at the same time,” an unredacted excerpt from a Microsoft document read.
“It is highly unusual that an end-product threatens the economic foundations of its essential suppliers, but that is the situation we have created for our LLM business with respect to its ‘content supply chain.’”
In a declaration submitted alongside its dueling summary judgment motion on Sept. 4, which was also refiled with fewer redactions on Thursday, another Microsoft exec framed Hecht as someone on the payroll to play a contrarian.
A company spokeswoman doubled down on distancing Microsoft from Hecht’s position in a statement to The News on Thursday.
“These comments reflect one employee’s individual perspective, are not a legal analysis, and do not represent the company’s views. Microsoft’s position is set out in its court filings, which explain why these transformative uses are consistent with copyright law and why Copilot is not a substitute for publishers’ journalism,” the statement read.
Regarding the “doom loop,” the spokesperson said the document concerned “broad principles and changes underway in how people find and consume information,” not copyright questions before the court.
The News also reached out to OpenAI for comment.
In their suit against the tech companies, the newspapers, among other allegations, argue that Microsoft’s Copilot and OpenAI’s ChatGPT are severely harming their business models by training large language models, or LLMs, on stolen stories, which are then regurgitated to users without credit, in violation of copyright laws. The Authors Guild and a host of best-selling writers are also suing the companies in the consolidated litigation.
The newspapers’ pending motion seeks summary judgment on the tech companies’ acquisition-, training-, and distribution-based infringement.
According to another portion of the motion that was unredacted on Thursday, on top of copying stories to teach LLMs to cook up content that looks a lot like news, the companies illicitly copied “billions of web pages” for “horse trading”-like deals — meaning they sold the content to each other rather than licensing it from the news organizations. Microsoft did not address a question about the allegation.
Figures unredacted from the filing refer to more than 3.9 million copies stolen from The Times for training and 7.3 million copies from The News and its sister papers, including The Chicago Tribune, Orlando Sentinel, Denver Post, and Mercury News. Sanctions are also being sought against OpenAI for allegedly destroying evidence and concealing its ability to locate stolen news stories in its framework.
The motion details that Microsoft recorded 83-93% drops in people even clicking on links to The Times, The News and its affiliated papers.
Other unredacted figures showed that survey takers who used to subscribe to The News and its sister papers were 28.6% more likely to ask A.I. for the news than people who still subscribe.
Regarding Times subscribers who use ChatGPT: 36% of survey respondents said using ChatGPT meant “I no longer need news from the New York Times at all,” unredacted data showed.
More sections of Thursday’s filings that were unredacted revealed that while public support necessary to produce reliable local journalism has reached a crisis point and there’s no lack of demand for solid reporting — millions of people are just asking ChatGPT for a summary instead of paying for subscriptions or reading a newspaper.
Microsoft’s experts have acknowledged in the suit that at least 1.3% of 45,000 Copilot conversations were directed to current affairs, according to another section of the unredacted motion.
The numbers were much larger for ChatGPT, which had 1 billion monthly users as of May, according to findings published this summer by Sensor Tower, a market intelligence and digital analytics platform.
“OpenAI’s internal data as of February 2026 tracked that ‘the demand for reliable local news is already visible inside ChatGPT at a rate of about 1 million prompts per week,’” one line unredacted on Thursday read.
The newspapers, in their summary judgment motion, said that the whopping figure is only the tip of the iceberg.
“While these figures reflect a remarkable number of news queries,” the filing read, “the evidence shows that the percentage of ChatGPT conversations seeking news is far higher.”
The tech behemoths contend their LLMs are taking the newspapers’ content and transforming it into something new, thereby constituting “fair use” and circumventing copyright laws.
The Trump Justice Department, which is not a party to the litigation, recently weighed in to cheerlead for OpenAI and Microsoft — saying the courts should side with the tech giants so America can win the global A.I. race.
The news outlets have repeatedly underscored in the litigation that, far from transforming content for educational benefits, LLMs often misinform users by distorting their news stories or providing “hallucinated” content in a highly confident tone, and by flooding the market with low-quality AI-generated “pink slime” news content farms.