“The Seattle Times and Newsday have filed a copyright infringement lawsuit against OpenAI and Microsoft, claiming their articles were used without permission to train AI models. The outlets also allege ChatGPT reproduces passages from their reporting verbatim in response to user queries. This adds to a growing wave of media litigation challenging the legal foundations of how large language models are built.”
Key Takeaways
- The Seattle Times and Newsday are the latest publishers to sue OpenAI and Microsoft for copyright infringement over AI training data.
- The outlets claim ChatGPT reproduces verbatim passages from their journalism when responding to user queries.
- The lawsuit mirrors similar legal actions already filed by other news organisations against OpenAI.
Two major US newspapers allege OpenAI stole their journalism to train ChatGPT.
trending_upWhy It Matters
As more news organisations pile into litigation against OpenAI and Microsoft, the courts are being forced to define whether scraping publicly available journalism for AI training constitutes fair use — a question with trillion-dollar implications. A ruling against OpenAI could force the company to renegotiate data licensing deals at scale or retrain models on narrower datasets, significantly raising operational costs. Publishers who have not yet struck licensing deals, like the Associated Press and Axel Springer have, may see this litigation as their primary leverage. Investors and developers building on top of OpenAI's API should watch these cases closely, as adverse rulings could reshape the underlying models they depend on.
FAQ
What exactly are the Seattle Times and Newsday alleging OpenAI did wrong?
They allege OpenAI used their published journalism without permission as training data for its AI models. They further claim that ChatGPT sometimes reproduces near-verbatim excerpts from their articles in response to user prompts, depriving them of traffic and revenue.
Is this lawsuit unique, or are other news outlets doing the same?
This is part of a broader wave of media litigation against OpenAI. The New York Times filed a landmark lawsuit in late 2023, and several other publishers have followed. The Seattle Times and Newsday cases represent a continuation of this coordinated industry pushback.
Could these lawsuits actually change how OpenAI builds its models?
If courts rule against OpenAI, it could be required to license training data from publishers, remove certain content from existing models, or pay damages. This would likely increase development costs and could set a legal precedent affecting the entire AI industry's approach to training data sourcing.
