Encyclopaedia Britannica and its Merriam‑Webster dictionary have sued OpenAI in Manhattan federal court, alleging the company trained ChatGPT on large swathes of their copyrighted material without permission and that the chatbot now reproduces or closely mirrors their content. According to the Independent, the complaint was filed on Friday by Britannica and its subsidiary, which say ChatGPT’s summaries have "cannibalized" web traffic that previously visited their sites.
The lawsuit asserts OpenAI copied nearly 100,000 encyclopedia articles, dictionary definitions and other online entries to develop its models, and that the system “generates outputs that copy or mimic, sometimes verbatim,” material from those sources. The companies argue the practice has deprived them of users and revenue and that the full extent of the alleged copying is known only to OpenAI.
Britannica and Merriam‑Webster also contend that OpenAI’s products misrepresent permission to republish their work and have even produced false attributions , so‑called AI “hallucinations” that cite Britannica in error. They are seeking unspecified monetary damages and a court order to stop further use of their content.
OpenAI responded by disputing the claims, saying its models are trained on publicly available data and that their use of such material is “grounded in fair use.” The company told reporters its technology “empower[s] innovation,” framing the case as part of a broader industry debate over how large language models are created.
Legal experts and publishers are watching the suit as part of a wave of litigation challenging how AI firms obtain training data. Last year, a group of authors reached a settlement with Anthropic after suing over similar claims; other publishers and creators have brought or threatened suits against AI developers on related grounds. Industry observers say the outcome could shape licensing expectations and business models for reference publishers and news organisations alike.
The Britannica case emphasises two battlegrounds: copyright law and the economic impact of generative AI on incumbent information providers. Britannica and Merriam‑Webster portray the dispute as defence of professionally produced, vetted reference content; OpenAI and some technologists argue that training on broad, publicly accessible corpora is central to the development of useful AI. The courts will now weigh those competing claims.
Source Reference Map
Inspired by headline at: [1]
Sources by paragraph:
- Paragraph 1: [2], [3]
- Paragraph 2: [3], [4]
- Paragraph 3: [4], [2]
- Paragraph 4: [1], [2]
- Paragraph 5: [1], [6]
- Paragraph 6: [5], [3]
Source: Noah Wire Services