Encyclopedia Britannica and Merriam-Webster have filed legal action against OpenAI, alleging in their complaint that the AI titan have gravely committed “massive copyright infringement”.
Britannica, which own Merriam-Webster, retain the copyright to nearly 100,000 online articles, which have been scraped and used to train OpenAI’s LLMs without permission, the publishing company allege in the lawsuit.
Britannica also accuse OpenAI of violating copyright laws when they generate outputs that contain “full or partial verbatim reproductions” of their content and when the AI lab uses their articles in ChatGPT’s RAG (retrieval augmented generation) workflow.
OpenAI’s RAG tool is how the LLM scans the web or other databases for newly updated information when responding to a query.
Britannica also allege that OpenAI violates the Lanham Act, a trademark statute, when they generate made-up hallucinations and attribute them falsely to the publisher – and which jeopardize “the public’s continued access to high-quality and trustworthy online information”.
“ChatGPT starves web publishers like [Britannica] of revenue by generating responses to users’ queries that substitute, and directly compete with, the content from publishers like [Britannica]”, the lawsuit reads.
Britannica join a number of other publishers and writers in pursuing legal proceedings against OpenAI over copyright issues. The New York Times, Ziff Davis (owner of Mashable, CNET, IGN, PC Mag, and others), and more than a dozen newspapers across the US and Canada, including the Chicago Tribune, the Denver Post, the Sun Sentinel, the Toronto Star, and the Canadian Broadcasting Corporation, have sued OpenAI.
There is not a strong legal precedent that establishes whether using copyrighted content to train an LLM is copyright infringement.
But in one particular instance, in June 2025, Anthropic successfully convinced federal judge William Alsup that this use case – using the content as training data – is transformative enough to be legal. However, Alsup argued that Anthropic violated the law by illegally downloading millions of books, rather than paying for them, which warranted a $1.5 billion class action settlement for impacted writers.
But many writers were unable to benefit due to specific eligibility requirements and procedural limitations set by the court and the settlement agreement (e.g., lack of US copyright registration, specific registration timelines, requirement of ISBN or ASIN, legal ownership and splits).
In any case, this marks the first time that the courts have given credence to AI companies’ claim that fair use doctrine can absolve AI companies from fault when they use copyrighted materials to train LLMs.
As dozens more cases over the relationship between AI and copyrighted works go to court, judges now have Bartz v. Anthropic to reference as a precedent. But given the ramifications of these decisions, maybe another judge will arrive at a different conclusion.
