The AI sector faces a new legal challenge, with The New York Times (NYT) filing a copyright infringement lawsuit against OpenAI, the developers of ChatGPT.
On December 27, The New York Times accused OpenAI in a lawsuit of illicitly utilizing its articles to train ChatGPT, alleging this has negatively impacted the Times' work. The lawsuit, referencing U.S. copyright laws and the Constitution, defends the newspaper's unique journalistic content and highlights instances of Microsoft's Bing AI using direct excerpts from its articles.
The lawsuit argues that OpenAI's tools distribute The New York Times' content without authorization, damaging its relationship with readers and leading to potential revenue losses from subscriptions, licenses, advertising, and affiliate partnerships. This issue isn't unique to The New York Times; in November, the News Media Alliance voiced similar concerns about AI chatbots infringing on copyrighted news, affecting the revenue, data, and user base of news organizations.
Cecilia Ziniti, an IP and AI attorney, called this the “strongest case yet” for copyright infringement against generative AI. She underscored the significance of The New York Times' website being a primary source for AI training, second only to Wikipedia and the U.S. patent document database. The lawsuit mentions that The New York Times approached Microsoft and OpenAI in April 2023 for a discussion on IP issues and potential resolutions, but talks were unsuccessful.
Ziniti, a decade-long subscriber to The New York Times and its Food Archive, stated on social media that if ChatGPT could provide her with free articles and recipes, she wouldn't need her subscription, suggesting the case could be pivotal for AI and copyright laws.
In the OpenAI developer community, the lawsuit sparked diverse reactions. Some users expressed hope that The Times' claim would fail, while others found the situation intriguing and worth The Times' effort. This lawsuit follows a similar pattern, with the Writers Guild of America suing OpenAI in September for using copyrighted material in AI model training.
OpenAI isn't alone in facing such legal issues. Universal Music Group sued Anthropic AI in October for allegedly infringing on copyrighted musical works. Additionally, a group of artists sued Midjourney, DeviantArt, and Stability AI for using their art in training image-generating AI models, with a U.S. judge initially dismissing the case due to insufficient evidence, before it was expanded with more comprehensive charges. OpenAI has pledged to bear all legal expenses for enterprise-level ChatGPT users entangled in copyright infringement lawsuits.
