Full Breakdown
OpenAI and Microsoft Face Copyright Lawsuit Over AI Training on News Content
By Drooid · · How we work
Core Legal Conflict: Fair-Use Defense Challenged by News Publishers
OpenAI and its primary investor Microsoft are defending a Manhattan federal-court lawsuit that accuses them of training large-language models on millions of newspaper articles without permission. The plaintiffs—beginning with *The New York Times* and now encompassing more than 550 publications—argue that the companies’ use of copyrighted material violates the Copyright Act and the Digital Millennium Copyright Act (DMCA). The defendants contend that the training process is “transformative” and therefore qualifies as fair use. Key filings were made public on September 4 and September 17, when the companies described their AI products as “substitutes for journalism.”
Background & Context
The suit was first filed in 2023 by a coalition of news outlets. Since then, the coalition has grown to include major papers such as the *Tampa Bay Times* (via Times Publishing Company), the *Austin Chronicle*, the Alternative Newsweekly Foundation (which supports more than 500 Black- and Latino-owned outlets), and the sports-news service SwimSwam. The plaintiffs are represented by Platkin LLP, a law firm founded by former New Jersey Attorney General Matthew J. Platkin. The Trump administration, in a filing on September 1, urged the court to view AI training as “extraordinarily” transformative.
Data & Statistics
- 550+ news organizations are party to the case, spanning local, regional, nonprofit, and specialty outlets.
- Internal documents cited in the lawsuit claim the training data set includes over 91,000 works from major news outlets.
- The plaintiffs allege that AI-driven search reduced traffic to *The New York Times* by up to 93 %, a figure presented as an internal estimate of revenue impact.
Official Statements & Responses
- OpenAI declined to comment on the newly unsealed filings.
- *The New York Times* spokesperson also declined to comment.
Conflicting Reports & Gaps
- Transformative use vs. substitution: OpenAI and Microsoft argue that AI training is transformative, while internal statements from company executives describe the products as “substitutive” and an “existential threat” to publishers.
- Company stance on employee remarks: Microsoft maintains that employee comments do not reflect official policy, whereas plaintiffs treat those remarks as evidence of intent.
Verbatim Quotes
- “Those observations should not be confused with conclusions about copyright questions before the court which Microsoft addresses in its filings.” — The spokesperson
The litigation remains active, and the court’s forthcoming rulings will shape how AI developers may incorporate copyrighted news material into future models.