The New York Times, OpenAI, and Microsoft filed dueling briefs Friday in a copyright lawsuit that could reshape how AI companies use published content to train their models. The filings, submitted in the U.S. District Court for the Southern District of New York, center on whether AI systems illegally reproduce news articles and whether those outputs damage the market for original journalism.
The Times, joined by other publishers, sued OpenAI and Microsoft in 2023, alleging the companies used millions of copyrighted articles without permission to build AI technologies that now compete directly with news outlets as information sources. Friday's briefs focused on two core legal questions: whether AI sufficiently transforms articles into new works, and whether AI-generated content serves as a market substitute that harms publishers' businesses.
The publishers' argument: preserving incentives for human creation
The Times argued that the economic viability of journalism depends on protecting copyrighted material from uncompensated use. "The future not just of journalism but of responsible AI, too, depends on preserving incentives for humans to produce the creative works on which a healthy society depends," the publisher wrote in its filing.
The news organizations contended that OpenAI and Microsoft cannot prove their products avoid harming the news market. The Times pointed to substantially lower click-through rates when traditional search links to its articles are replaced by generative AI summaries in search results. That substitution effect, the publishers argued, demonstrates that the AI systems do not sufficiently transform the works into new forms of expression - a key test in fair use analysis.
For legal professionals tracking how courts apply AI for Legal precedent, the case raises threshold questions about transformation and market harm under copyright law.
The tech companies' defense: copyright law does not block transformative technology
Microsoft argued that existing copyright law "does not permit rights holders to block transformative technologies" such as its Copilot product. OpenAI made a parallel argument, writing that "these technologies put the world's facts within reach of anyone who asks, in whatever form is most useful to them. That is progress, and copyright law does not stand in its way."
OpenAI cited expert witnesses who said the Times' web traffic was "not negatively impacted" by ChatGPT, directly contradicting the publishers' claims of market harm. The company also invoked a 2023 Supreme Court ruling involving Andy Warhol, who was found to have infringed a photographer's copyright on an image of Prince. The Court ruled Warhol's silk-screened version did not constitute fair use - a precedent the publishers have leaned on in their arguments.
Cultural and historical references in the briefs
The filings ranged beyond legal doctrine into literature, art, and sports. Microsoft invoked Lord Byron, the 19th-century British poet who believed "literature and science were at odds," and noted that Byron's daughter, Ada Lovelace, is recognized as the world's first computer scientist. Lovelace, Microsoft wrote, saw early computers as examples of "poetical science" - a framing the company applied to modern AI programs.
OpenAI, making a point about the breadth of its technology while nodding to the court's local history, referenced a web search for the New York Giants' 1951 National League pennant. The team lost that year's World Series to the Yankees.
Why this matters for legal professionals
The case is one of the most closely watched AI copyright disputes, but it remains in early stages. According to a tracker maintained by Ed Lee, a professor at Santa Clara University School of Law, most AI copyright lawsuits have not gone to trial, and only a handful have produced rulings on key legal issues. The outcome here will likely influence how courts assess fair use claims for training data, the standard for proving market harm in AI-related infringement cases, and whether AI-generated summaries constitute transformative use. For paralegals and attorneys handling document review or IP matters, the case offers a live example of how traditional copyright frameworks are being stress-tested by generative AI - a topic covered in depth through AI for Paralegals training resources.
Your membership also unlocks:
