OpenAI pushes back against order to hand over millions of ChatGPT convos

share on

twitter /
facebook /
linkedin /
- email
- telegram
- whatsapp
- wechat
- pinterest
- line
- snapchat
- reddit

OpenAI is pushing back against a US court order requiring it to hand over 20 million anonymised ChatGPT conversations to The New York Times and other publishers as part of an ongoing copyright infringement lawsuit.

In a filing on Wednesday (13 November), the company reportedly asked a federal judge in New York to reverse the order, arguing that producing the chat logs would expose confidential user information. OpenAI claimed that “99.99%” of the transcripts have no connection to the copyright claims raised in the case.

OpenAI warned in its filing that the order could mean any ChatGPT user from the past three years might see their personal conversations handed over to The New York Times for review as part of the case, according to Reuters.

Don't miss: OpenAI’s APAC comms head on leveraging ChatGPT as a strategic partner

The New York Times and other outlets reportedly maintain that the data is crucial to determine whether ChatGPT reproduced their copyrighted material and to challenge OpenAI’s accusation that they “hacked” the chatbot’s responses to create evidence. The lawsuit alleges OpenAI unlawfully used the publishers’ articles to train its large language models.

Magistrate Judge Ona Wang, who previously ruled in favour of disclosure, reportedly said the chat data would undergo “exhaustive de-identification” and other safeguards to protect user privacy. OpenAI faces a Friday deadline to comply with the order.

In a blog post, Dane Stuckey, chief information security officer at OpenAI, condemned the court order, describing it as “an invasion of user privacy”.

“The New York Times is demanding that we turn over 20 million of your private ChatGPT conversations,” he said. “This demand disregards long-standing privacy protections, breaks with common-sense security practices, and would force us to turn over tens of millions of highly personal conversations from people who have no connection to The Times’ baseless lawsuit.”

Stuckey said the company had previously resisted broader demands from The New York Times to access 1.4 billion chat logs and to restrict users’ ability to delete private conversations. He added that OpenAI had proposed “privacy-preserving alternatives”, such as targeted searches for chats containing NYT text, but those were rejected by the publisher.

The company also reiterated that its enterprise, education, business, and API customers are not affected by the court order, which applies only to a random sample of consumer ChatGPT conversations from December 2022 to November 2024.

According to Reuters, a New York Times spokesperson countered that OpenAI’s post misrepresented the situation, noting that user privacy would remain protected since the court had only ordered the release of anonymised chat samples under a legal protective order.

MARKETING-INTERACTIVE has reached out for more information.

The OpenAI lawsuit is part of a growing wave of legal challenges testing how generative AI companies source data to train their systems, and where copyright boundaries lie. Earlier this month, the long-running legal clash between Getty Images and Stability AI became one of the first major test cases for copyright in the generative AI era. According to reports, Getty largely lost its London lawsuit accusing Stability AI of unlawfully using its copyrighted photos to train the image generator Stable Diffusion.

Judge Joanna Smith ruled that Getty succeeded only “in part” on trademark infringement, specifically in relation to Getty watermarks reproduced by users of Stable Diffusion, while dismissing broader copyright claims. The case underscored how courts are still grappling with whether training AI models on copyrighted works constitutes infringement, an issue now central to OpenAI’s own legal fight with The New York Times.

share on