Encyclopaedia Britannica sues OpenAI for intellectual property theft

Encyclopaedia Britannica, one of the world's oldest and most reputable encyclopaedias, has filed a lawsuit against OpenAI, accusing the artificial intelligence company of using knowledge accumulated over decades without permission. The dispute raises a broader question: is AI trained unfairly using intellectual property created by others? The outcome of this case could influence the future of AI development more broadly.

Encyclopaedia Britannica, one of the world's oldest and most esteemed sources of knowledge, has filed a lawsuit against OpenAI, accusing the artificial intelligence company of using its intellectual property without authorisation. Britannica claims that OpenAI has used materials accumulated over decades to train its models without permission or compensation.

The dispute between AI and copyright

At the heart of the lawsuit is a question about what AI models actually do: do chatbots create new knowledge or repackage someone else's work? According to Britannica, the answer is the latter, and OpenAI's use constitutes essentially theft-the fruits of decades of dedicated work have been turned into a commercial product without the consent of the original authors and publishers.

This is not the first dispute of its kind. In recent years, multiple publishers, newspapers, and authors have filed similar claims against AI companies. The question has also resonated in US courts, where the debate centres on whether data mining used in training large language models constitutes copyright infringement.

Broader implications for the future

The Britannica case stands out from others, chiefly because of the company's weight and standing-it is an institution over 250 years old with strong moral authority in the preservation and dissemination of knowledge. The outcome of this lawsuit could serve as a precedent for shaping the terms and conditions under which AI models will be trained in the future.

If the court sides with Britannica, it could force AI companies to negotiate licensing agreements with intellectual property owners and pay for content use. OpenAI has previously emphasised in similar disputes that training a model is comparable to human reading and does not constitute copyright infringement-a claim that courts have neither definitively accepted nor rejected.