OpenAI Copyright Trial: New York Times Alleges Hidden Evidence in ChatGPT Lawsuit

<img src="/wp-content/uploads/2025/11/educate_regulate_adapt_policies_6batc.jpg" alt="OpenAI copyright trial and New York Times hidden evidence allegations" style="width:100%;border-radius:12px;margin-bottom:2rem;">

The New York Times and The Daily News claim that OpenAI has been lying about its ability to search customer chat log data and training datasets for their copyrighted works. It is the latest escalation in a two-year lawsuit against the AI firm for allegedly violating copyright law by training its generative AI models on the Times' content and reproducing that journalism in user outputs.

According to reporting by TechCrunch on July 9, 2026, the outlets allege that OpenAI had already conducted internal searches of its training corpus to locate copyrighted journalism works before the Times filed suit. They further claim OpenAI assembled a database of about 78 million de-identified ChatGPT conversations for internal use, and implemented a so-called Bloom filter as part of a set of tools called Project Giraffe to detect regurgitation in outputs shortly after the lawsuit was filed.

What the plaintiffs are asking the court to do

The Times and The Daily News are now asking the judge to discipline OpenAI for allegedly withholding evidence and interfering with the discovery process. Their requests include preventing OpenAI from using the 20 million chat log sample it submitted as evidence, because they claim the sample is unreliable. They also want the court to accept as fact that ChatGPT logs would have shown major regurgitation and grounding of the plaintiffs' content, and to prevent OpenAI from arguing that its provided chat logs do not demonstrate substantial regurgitation. Finally, they are asking the court to make OpenAI pay legal fees for the effort required to chase down this evidence.

Ian B. Crosby, lead counsel for the plaintiffs, said in a statement: "If OpenAI genuinely believed that copying our clients' journalism was fair and legal, it wouldn't have hid the truth about having done it."

Why this matters for authors

The case is the clearest signal yet that the legal line around AI training data is still being drawn, and that the process is messy. For authors who rely on AI tools for drafting, research, or editing, the central question is not whether models are useful. It is whether the companies behind those models have respected the copyrights of the writers whose work helped train them.

The trial also has a practical dimension. If courts find that OpenAI deleted data after litigation began, or substituted logs in a court-ordered sample, that changes how reasonable observers should treat the output of any AI system trained on similar data. Authors who use AI-generated content in their own books should keep records of how that content was produced, revised, and sourced. That documentation matters more when the underlying model's training history is under scrutiny.

OpenAI spokesperson Drew Pusateri denied the allegations in a statement to TechCrunch. He accused the Times of trying to access private user conversations as its case weakens. "As the Times' case weakens and they've been forced to drop claims against us, they're persisting with their efforts to invade the privacy of people who have nothing to do with this case, including by making these blatantly false allegations," Pusateri said. "We'll continue defending our users' privacy and the long-established principles of fair use."

What to watch next

The judge's decision on the sanctions motion could narrow or expand what OpenAI can argue about its own training data. A ruling that accepts the plaintiffs' factual assertions would make it harder for OpenAI to claim that its models do not reproduce copyrighted material. That, in turn, could affect how authors think about using AI-generated prose in published work, especially when that prose resembles existing journalism or literature.

For now, the safest path is to treat AI as a drafting partner rather than a source of publishable text. Use it for outlines, scene ideas, or first drafts, then rewrite in your own voice and keep a clear record of what changed. If you want a workflow that keeps that record intact from the first chapter through the final export, you can start a free trial and see whether persistent context changes how one chapter flows into the next.