Dictionary Takes Legal Action Against OpenAI
Encyclopedia Britannica and Merriam-Webster have initiated legal proceedings against OpenAI, accusing the AI leader of significant copyright violations. The lawsuit claims that OpenAI’s language models utilized nearly 100,000 copyrighted articles without proper authorization.
Details of the Lawsuit Against OpenAI
According to the legal complaint, Britannica asserts that OpenAI has infringed on its copyright by generating outputs that reproduce content—either in full or in part—verbatim. Additionally, the complaint alleges violations of copyright law within the workflow of ChatGPT’s Retrieval Augmented Generation (RAG) tool, which helps the model gather updated information from various sources.
Allegations of Trademark Violations
Britannica also claims that OpenAI violates the Lanham Act, which governs trademarks. This occurs when ChatGPT creates fictional responses and inaccurately attributes them to Britannica. Such practices reportedly undermine the integrity and credibility of high-quality online information.
Impact on Revenue for Publishers
- The lawsuit states that ChatGPT threatens revenue streams for publishers like Britannica.
- OpenAI’s responses may compete directly with original content from publishers.
Britannica is not the only entity to take legal action. Other notable plaintiffs include The New York Times, Ziff Davis, and over a dozen newspapers in the U.S. and Canada. This growing coalition highlights a widespread concern about the effects of AI on published content.
Legal Precedents and Implications
While there is limited legal precedent regarding the use of copyrighted material for training language models, a past case involving Anthropic could offer some insights. A federal judge ruled that using copyrighted content in a transformative manner could be permissible. However, Anthropic faced consequences for illegally downloading books, leading to a substantial $1.5 billion settlement for affected authors.
Conclusion
As the legal landscape surrounding AI and copyright continues to evolve, the action taken by Britannica and Merriam-Webster against OpenAI may set important precedents. The outcomes could significantly impact how artificial intelligence interacts with existing published works.
