Encyclopaedia Britannica sues OpenAI for massive copyright infringement

TL;DR: Encyclopaedia Britannica and its subsidiary Merriam-Webster have filed a copyright lawsuit against OpenAI. The publisher claims nearly 100,000 online articles were scraped without permission to train ChatGPT, and accuses OpenAI of falsely attributing hallucinated content to Britannica.

The lawsuit targets three distinct alleged violations: using copyrighted articles as training data, generating outputs that contain verbatim reproductions of Britannica content, and using the publisher’s articles in ChatGPT’s retrieval augmented generation (RAG) workflow, which scans the web for current information when responding to queries.

A trademark twist

Beyond copyright, Britannica alleges OpenAI violates the Lanham Act, a US trademark statute, when ChatGPT generates fabricated information and attributes it to the publisher. This legal angle goes further than most AI copyright cases by arguing that hallucinations actively damage Britannica’s reputation for accuracy.

“ChatGPT starves web publishers like [Britannica] of revenue by generating responses to users’ queries that substitute, and directly compete with, the content from publishers,” the lawsuit states.

Growing publisher backlash

Britannica joins a lengthening list of publishers suing OpenAI. The New York Times, Ziff Davis (owner of Mashable, CNET, and IGN), and newspapers across the US and Canada including the Chicago Tribune and Toronto Star have all filed similar claims. A separate Britannica lawsuit against Perplexity is also pending.

Legal precedent remains unsettled. In one notable case, a federal judge ruled that Anthropic’s use of copyrighted books as training data was transformative enough to be legal, but that the company violated the law by downloading millions of books without paying for them. That case resulted in a $1.5bn (roughly £1.1bn) class action settlement.

Looking forward

For UK publishers watching these proceedings, the outcome could shape how AI firms are permitted to use copyrighted content globally. British media organisations face similar questions about whether AI training constitutes fair use or commercial exploitation, and the Britannica case’s trademark angle offers a new line of attack that European publishers may follow.