The New York Times and three Tribune papers sued Brave, saying it scraped news for its search index and sold access to AI firms like Anthropic.
The Brave lawsuit filed in Manhattan federal court on Monday, November 2, accuses Brave Software of scraping news articles into its search index and selling access to that content to AI companies including Anthropic, Perplexity, Mistral, xAI and Cohere. The New York Times filed one complaint, and the New York Daily News, Chicago Tribune and Orlando Sentinel filed a second, closely matching one in the same court.
Both cases target three Brave products: the Ask Brave AI answer feature in Brave Search, the paid Brave Search API, and the Brave browser itself. None of the AI companies named in the filings is a defendant, and none of the claims has been tested in court yet.
What the Brave Lawsuit Claims
According to the 72-page Times complaintBrave built an index of “over 30 billion webpages” without ever disclosing the name of the crawler that collects them, which the paper says leaves site owners with no way to block it. The Times calls Brave “the supplier, dealer, and distributor of stolen copyrighted content” for its own users and for AI developers.
The filing says Ask Brave, launched in September 2025, answers queries with “verbatim text from and detailed summaries” of articles, while the Search API lets other companies use the index to train models or ground chatbot answers in real time. It also cites a Brave blog post stating that Anthropic used the Search API while training Claude 2. Citing third-party estimates, the Times says Brave has made “more than $100 million” from publishers’ work, while “publishers have received nothing.”
The complaint puts Brave’s reach at close to 130 million monthly browser users as of September 30, 2026, up from the 101 million users the company reported earlier, with Brave Search handling more than 50 million queries a day.
Paywall Blocker and Ad Swapping
The publishers also go after the browser’s “Paywall blocker” setting, which Brave promotes as a way to “Bypass Paywalls.” The Times says this violates the DMCA’s anti-circumvention rules. Ars Technica tested the Mac browser and was able to read a paywalled Times article with the feature, which was on in the default settings and can be switched off.
Another complaint targets advertising: Brave blocks ads on news sites by default but shows its own Brave Ads and shares revenue with users. The Times argues this takes ad income that pays for its journalism.
What the Publishers Want
The Times brings five counts, including direct, contributory and vicarious copyright infringement, DMCA violations and unjust enrichment. It asks for a permanent injunction, statutory and actual damages, and disgorgement of the profits Brave earned. The paper says it sent cease-and-desist letters to Brave CEO Brendan Eich in August 2025 and April 2026 and got no reply.
The Tribune-owned papers say more than 2 million registered articles are at stake in their case. Their lawyer, Steven Lieberman, said, “Buying stolen goods secondhand doesn’t make them legitimate.”
Brave had not publicly responded as of November 3. Perplexity, which the Times is already suing separately, called the filing “disinformation” without pointing to specific errors.
Why It Matters
The case shifts the fight over AI training data from model makers to the companies that supply them with web data. Search APIs like Brave’s have become a common way for chatbots to fetch fresh information, so a ruling against Brave could raise costs for many AI products. It also adds pressure to industry talks on an AI training opt-out standard, and follows earlier suits such as one from a New York Times reporter against AI companies.