News Corp countersues Brave over alleged AI article scraping

News Corp countersued Brave on July 22, accusing the search and browsing company of scraping articles for artificial intelligence use. The filing pushed a California dispute deeper into one of the most consequential fights in media and technology: whether companies can ingest, summarize or repurpose journalism at scale without permission or payment.

The clash began in March 2025, when Brave Software sued News Corp in the U.S. District Court for the Northern District of California, case 3:25-cv-02503. Brave sought a declaratory judgment that indexing News Corp content was not copyright infringement and that its use with AI chatbots was lawful after News Corp sent a cease-and-desist letter. Judge Jon S. Tigar is overseeing the Northern District litigation.

At stake is the line between ordinary search and automated extraction. Search indexing generally crawls public webpages so users can locate stories and click through to the original publisher. Scraping, by contrast, usually refers to automated copying or collection of article text, headlines and other copyrighted material, often at scale, for use in AI products that can answer questions directly or repurpose the material inside a product. News Corp’s new complaint signals that it believes Brave crossed that line.

Brave has positioned itself as a privacy-minded, AI-forward alternative to larger search products, and it has argued that bundling copyrighted articles that can be licensed and sold is not unlawful. News Corp, by reversing the posture and filing its own case, is pressing the opposite theory: that a tech company cannot build a valuable AI service on the back of publisher reporting without a license. The legal questions now include not only infringement, but market harm, fair use and whether AI outputs substitute for the original article.

The business consequences are immediate. News Corp owns a wide portfolio of news brands and has pushed for compensation from AI firms that rely on premium journalism. If News Corp wins, newspapers, magazines and wire services would likely gain leverage in licensing talks, and AI product design could shift toward explicit permissions, paid data pipelines and tighter controls on what gets ingested. If Brave wins, large-scale web collection would become easier to defend, making it harder for publishers to police how their work is indexed, summarized or reused.

The dispute also fits a broader wave of copyright battles involving publishers, model makers and search companies. A separate Northern District of California case, 3:26-cv-04758, involving Brave Software and News Corporation was filed on May 19, 2026. The U.S. Copyright Office’s pre-publication version of Part 3 of its report on generative AI training, released in May 2025, shows how unsettled the law remains over training data, market harm and fair use.