USA Today Co. is seeking more than US$250 million in a copyright lawsuit against OpenAI, alleging that its journalism was copied without authorization to train AI models. The case adds another major publisher to the disputes over how AI companies obtain and use news content.
For media businesses and marketers, the filing points to an unresolved commercial boundary: making journalism accessible to AI systems does not, by itself, settle the terms under which that work can be used. The allegations have not been adjudicated, and the damages sought are not compensation already awarded.
Key Takeaways
- USA Today alleges that OpenAI used its journalism to train models without permission.
- The lawsuit’s damages request remains a claim, with no ruling established in the reviewed reporting.
- AI-friendly publishing formats and permission to use content are separate commercial decisions.
Table of contents
Jump to each section:
- What USA Today is seeking
- Licensing and litigation can coexist
- The practical signal for content businesses
What USA Today is seeking
More than US$250 million in damages sought. This is the amount requested in USA Today’s new lawsuit, according to reporting on the filing; it is not a court award.
USA Today and several of its newspaper businesses filed the suit in New York on October 8. The publisher alleges that OpenAI copied hundreds of thousands of articles to train its models without permission, including reporting from local titles such as The Tennessean, IndyStar, The Columbus Dispatch and The Oklahoman.
The complaint frames that use as a commercial problem for the businesses that financed the reporting. Producing original journalism requires staff, editing and distribution; allowing another service to incorporate that work without an agreement raises a separate question about who captures its value.
Those are the plaintiffs’ claims. The filing does not establish infringement, and the requested damages are not an award. OpenAI had not responded to a request for comment by the time the reporting used for this article was published. No later response was verified for this article.
The distinction matters for media buyers and content owners. A large damages demand signals the seriousness of a dispute, but it does not tell them what a court will decide or what a future licensing agreement might cost.
Licensing and litigation can coexist
USA Today’s dispute sits alongside its interest in making journalism available to AI companies through commercial agreements. ContentGrip previously covered its experiments with machine-friendly article formats, a different development from this lawsuit. The two activities address different parts of the same business question: how publishers can set terms for the use of their work.
Preparing content for machine consumption is a distribution decision. Granting permission to use it is a rights decision. A cleaner format can make an article easier to ingest, but that alone does not determine whether an AI company may train on it, retrieve it or reproduce passages for users.
Other publishers, including The New York Times, The Intercept and Ziff Davis, have also pursued copyright claims against OpenAI. Their presence provides context for USA Today’s decision, although the complaints and potential outcomes are separate. It would be premature to treat the latest filing as a settled industry standard.
For publishers exploring AI revenue, that leaves licensing discussions and rights enforcement running in parallel. A commercial relationship with one AI provider need not imply permission for every provider, every product or every form of content use.
The practical signal for content businesses
USA Today is a news publisher with a local business marketing operation through LocaliQ. That combination makes the case relevant beyond national news brands: original local reporting can also be an input into products that answer questions or summarize information for an audience.
For marketers commissioning research, interviews or branded editorial work, the useful implication is to distinguish ownership from usage permission. Contracts with contributors, publishers and technology vendors can describe who holds the rights and which uses are permitted. An AI supplier’s access to material is a separate issue from a brand’s permission to republish it.
Teams evaluating an AI content tool can ask how it obtains source material, whether it has relevant licenses and how it handles attribution and rights complaints. Those questions do not require predicting this lawsuit’s outcome. They help establish what a vendor can actually substantiate about its inputs.
The case also gives media businesses a reason to keep their commercial plans specific. An AI partnership should spell out the licensed content and uses rather than treating visibility in AI answers as interchangeable with training rights. Whether USA Today wins its claims remains unresolved; the immediate business signal is that the publisher is actively contesting unauthorized use while looking for ways to monetize authorized access.
