USA Today Co Sues OpenAI for $250M Over Training Data

USA Today Co. is suing OpenAI, alleging the company copied hundreds of thousands of its articles to train AI models without authorization. The publisher seeks more than $250 million in damages The Verge.

The price tag is explicit. The core claim is simple. OpenAI never asked for permission to use the content, according to the complaint.

The suit encompasses USA Today Co. properties beyond the flagship title, including The Tennessean, Indy Star, The Columbus Dispatch and The Oklahoman. That structure matters for technologists to understand. The allegation is not about a single crawl or a single output. It describes systematic ingestion of a large, continuously updated news corpus into training pipelines for general-purpose models.

USA Today Co. is not litigating in isolation. The New York Times copyright lawsuit against OpenAI and Microsoft was first filed in 2023 and accuses the companies of using millions of newspaper articles. Microsoft is OpenAI's largest financial backer Reuters.

A separate action from The Seattle Times and Newsday also names OpenAI and Microsoft as defendants for alleged copyright infringement. That case was filed in the U.S. District Court for the Southern District of New York and alleges the companies scraped the newspapers' websites Reuters.

The docket keeps expanding. More than 30 U.S. local newspapers that together own about 400 titles have sued OpenAI and Microsoft over unauthorized scraping and use of their content Press Gazette. An earlier eight-newspaper action accused ChatGPT-maker OpenAI of copyright infringement, alleging the technology companies purloined millions of copyrighted news articles without permission or payment.

The dispute extends beyond daily news. Encyclopedia Britannica alleged that OpenAI unlawfully copied nearly 100,000 of its articles to train GPT large language models. Major publishers have separately alleged that Meta used millions of books and articles without permission to train its Llama AI model. Authors suing Microsoft over an AI model allegedly trained on pirated books requested a court order blocking infringement and statutory damages of up to $150,000 for each work allegedly misused.

For practitioners, the technical questions are consistent across these cases. How was the corpus acquired. Whether web-scale crawling, dataset curation, tokenization and weight updates constitute reproduction or transformation under copyright law. Where the line sits between training-time ingestion and inference-time memorization or regurgitation. None of those questions turn on model architecture alone. They turn on provenance, logging and control.

The broader context here is that courts are being asked to define the rules for the data supply chain that underpins foundation models. Publishers argue for prior consent and compensation. Model developers have generally pointed to fair use and the transformative nature of training. In this author's view, that binary framing obscures the more durable engineering and business problem.

Worth flagging is the operational gap. The web was built for indexing and retrieval, with robots.txt and terms of service as imperfect proxies for permission. Large-scale training broke those assumptions. It requires machine-readable rights signaling, durable provenance metadata and licensing interfaces that work at crawl scale, not through one-off negotiations.

Looking at what this means for builders and buyers, the near-term risk is less about any single verdict than about fragmentation. Different publishers, different jurisdictions, different licensing demands. Enterprise teams already manage data governance, model cards and indemnification clauses. That discipline will now have to extend upstream to training data sourcing and downstream to output filtering.

The long arc remains constructive. Structured access to high-quality editorial, reference and book content would improve models. It would reduce hallucination, improve citation and create a revenue path for original reporting and research. My own children grew up moving from printed encyclopedias to Wikipedia to chatbots, and each shift made information easier to get while making its source harder to see. Fixing that provenance layer is, in this author's view, where the real opportunity lies.