Key Takeaways

• OpenAI is contesting a court order requiring it to retain all user data, including deleted chats, in response to a lawsuit by The New York Times.

• The lawsuit claims that OpenAI and Microsoft used certain content without permission to train AI models, violating copyright law.

• The central legal question revolves around whether AI training constitutes “fair use” or copyright infringement.

• OpenAI asserts that preserving all user data poses a serious privacy risk and oversteps legal bounds.

• The case is part of a growing trend of copyright lawsuits targeting major AI companies, including similar cases filed by Ziff Davis, Reddit, and music publishers.

Why OpenAI Is Under Legal Pressure

The dispute arises from a May 13th federal court directive requiring OpenAI to “preserve and segregate all output log data that would otherwise be deleted on a going forward basis until further order of the Court.” This order could have massive implications for how user data is stored, especially for services like ChatGPT, where users regularly delete chat histories.

OpenAI’s Response To The Lawsuit

Source: OpenAI

The legal battle began in December 2023, when The New York Times filed a lawsuit against OpenAI and Microsoft, accusing them of unlawfully using its copyrighted content to train AI models, including ChatGPT and Microsoft’s Bing Chat.

The Core Legal Issue

The alleged unauthorized use of the content not only infringes upon various intellectual property rights but also poses a threat to the traditional journalism business model. There is concern that evidence of potential copyright violations could be lost as users continue to erase their conversations with AI tools.

At the center of the lawsuit is the legal interpretation of “fair use”, a longstanding doctrine that allows limited use of copyrighted material without permission under certain conditions, such as commentary, criticism, or education.

Fair Use vs. Copyright Infringement

The New York Times argues that OpenAI’s models reproduce near-verbatim excerpts from its articles and, at times, circumvent paywalls by generating AI-written summaries. If proven true, these actions could severely undermine the publication’s content monetization strategies and brand value.

In contrast, OpenAI maintains that its use of publicly available data falls within the bounds of fair use, asserting that the content is transformed through training rather than directly reproduced.

Moral High Ground & Public Opinion

Both parties claim to be acting in defense of larger principles.

• The New York Times says it’s standing up for journalistic integrity and sustainability, emphasizing that news organizations must be paid fairly when their work is used to build commercial products.
• OpenAI CEO Sam Altman has publicly accused the newspaper of being “on the wrong side of history,” asserting that AI technology can benefit humanity when developed responsibly.

The company also criticized The New York Times for what it described as “cherry-picking data” in its legal arguments, claiming that the examples cited do not represent typical model behavior.

AI, Copyright, & The Expanding Legal Landscape

As generative AI becomes more embedded in daily life, legal challenges surrounding data use, privacy, and copyright are becoming increasingly common. Courts are now pivotal in deciding how AI companies can train their models, and what kind of data is fair game.

The lawsuit from The New York Times is far from an isolated case. In April 2025, Ziff Davis, owner of outlets like PCMag and Mashable, also filed suit against OpenAI, alleging the misuse of its media content in AI training.

Reddit Clashes With Anthropic

Just this week, Reddit filed a lawsuit against Anthropic, another AI company, accusing it of scraping Reddit data without proper licensing. Anthropic is also under fire from music publishers and authors who claim it used protected works without authorization.

Reddit’s Lawsuit Against Anthropic

Source: Reddit

In any case, these lawsuits indicate a broader industry reckoning as creators, tech firms, and courts wrestle over how intellectual property laws should apply in the age of generative AI.

FAQ

Why is OpenAI being sued by The New York Times?

The Times alleges that OpenAI used its copyrighted content to train ChatGPT without permission, which could harm its business and violate intellectual property laws.

What does the court order require OpenAI to do?

The court has instructed OpenAI to preserve all user data, including deleted chats, to prevent potential evidence from being lost during the lawsuit.

Is OpenAI complying with the court order?

OpenAI is currently appealing the decision, arguing that the order is an overreach and threatens user privacy.

What is “fair use” and how does it apply here?

Fair use allows for limited use of copyrighted material without permission for purposes like commentary, news reporting, or education. The case will hinge on whether training AI models qualifies under this doctrine.

Are other companies facing similar lawsuits?

Yes. Companies like Anthropic and OpenAI are facing lawsuits from media outlets, music publishers, and authors over the unauthorized use of content for AI training.