Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Presented in New York on March 13, 2026, the complaint accuses OpenAI of using reference publishers’ content as AI training data without permission and then generating responses that reproduce it verbatim, six months after the same companies sued Perplexity for almost identical reasons.
Encyclopedia Britannica and Merriam-Webster have filed a copyright and trademark lawsuit against OpenAI in the U.S. District Court for the Southern District of New York, alleging that ChatGPT has been trained and continues to reproduce their copyrighted content without authorization, to the material detriment of both publishers.
The complaint, filed on March 13, 2026 (Case No. 1:2026cv02097), accuses OpenAI of using nearly 100,000 Britannica online articles as training inputs for its AI language models. The full extent of the copying, the complaint acknowledges, is known only to OpenAI itself.
Britannica, which owns Merriam-Webster as a subsidiary, argues that the law does not permit OpenAI’s systematic disregard for its intellectual property rights and holds it accountable for the substantial harm it is causing and the profits it is reaping through that infringement.
The 💜 of EU technology
The latest rumors from the EU tech scene, a story from our wise founder Boris and some questionable AI art. It’s free, every week, in your inbox. Register now!
“ChatGPT then provides narrative responses to user queries that often contain verbatim or near-word-for-word reproductions, summaries, or summaries of the original content, including plaintiffs’ copyrighted works.” – of the complaint.
The lawsuit is structured around two legal pillars, both of which mirror the framework that the same plaintiffs used when they sued the artificial intelligence search engine Perplexity in September 2025. The first is copyright infringement under the Copyright Act of 1976, which gives authors the exclusive right to reproduce and distribute their works.
Britannica argues that OpenAI violated those rights at multiple stages: by scraping its websites to create training inputs, by feeding that content into its models during training, and then by generating results that closely reproduce or summarize the originals when users query ChatGPT on topics covered by Britannica’s editorial catalog.
The second pillar is trademark law under the Lanham Act. By presenting AI-generated responses, which may contain inaccuracies or hallucinations, along with Britannica and Merriam-Webster’s famous trademarks and brand identities, the complaint argues that OpenAI misleads users into believing that Britannica or Merriam-Webster has endorsed or is the source of those responses.
Britannica’s reputation is built on precision built over more than 250 years; Associating that brand with information manufactured by AI, the complaint maintains, causes direct reputational damage that goes beyond the mere loss of copyright.
The business model under pressure
The underlying business case follows the logic laid out in the broader wave of publisher lawsuits against AI companies. Britannica’s current business is primarily digital and relies on subscriptions and advertising revenue that depends on web traffic.
When ChatGPT answers a user’s question about, for example, the causes of the French Revolution or the properties of a chemical element using content from Britannica articles, those users have fewer reasons to visit the Britannica website directly. The complaint describes ChatGPT as a free user of Britannica’s high-quality, trusted content, transferring the value of that content to OpenAI without compensation.
Britannica discontinued its 32-volume print edition in 2012 and went entirely digital. That transition made the quality and exclusivity of its editorial content, produced by researchers, writers and editors working to the standards that the brand has maintained since its founding in Edinburgh in 1768, the core asset of the business.
The complaint frames OpenAI’s alleged copying not simply as a legal violation but as an existential threat to a model that cannot survive if the economic benefits of that content flow to AI platforms rather than their creators.
The Perplexity Precedent and What Comes Next
This is not the first case of its kind at Britannica. In September 2025, the same plaintiffs, Encyclopedia Britannica and Merriam-Webster, filed an essentially parallel complaint against Perplexitythe AI-powered answer engine.
That complaint, also filed with the SDNY, alleged that Perplexity’s system scraped Britannica content to create its responses in real time, bypassing robots.txt protections and presenting verbatim or near-word-for-word reproductions under the guise of AI-generated summaries. The Perplexity case is still ongoing.
The OpenAI case is structurally similar, but reaches into a significantly more complex legal landscape. OpenAI is already the subject of a large multidistrict litigation (MDL) at the SDNY, currently overseen by Judge Sidney Stein, consolidating more than a dozen copyright lawsuits brought by news publishers, including the New York Times.
That MDL is nearing the close of fact discovery, and a fair use resolution is not expected before summer 2026 at the earliest. Analysts following the litigation landscape, including legal tracker ChatGPT Is Eating the World, have noted that the Britannica-OpenAI case will most likely be transferred to that MDL and then remain pending its outcome, meaning a resolution on the merits could take years.
For now, the complaint brings the total number of copyright lawsuits filed against artificial intelligence companies in the United States to 91, according to the same tracker. OpenAI had not publicly responded to the complaint at the time of writing.
The Britannica filing lands in a market where legal and licensing approaches to AI content are diverging sharply. At the same time as publishers are filing lawsuits, a growing number of media organizations have signed licensing agreements with artificial intelligence companies; News Corp signed a deal with Meta worth up to $50 million a year in March 2026; British publisher Reach struck a usage-based deal with Amazon for its Nova AI model the same month.
The Anthropic copyright case (Bartz v. Anthropic), involving the use of pirated books for AI training, reached a $1.5 billion class action settlement in 2025, the largest in the wave of AI copyright litigation so far, establishing that these cases can produce significant financial consequences.
Whether Britannica’s lawsuit against OpenAI follows the path of a settlement, trial or eventual consolidation in the MDL depends in part on the trajectory of the MDL itself. What the presentation makes clear is that publishers with strong brand equity and a defensible claim to accuracy, the encyclopedias and dictionaries whose authority AI systems are arguably borrowing and monetizing, are not content to wait for the industry to self-regulate.
