OpenAI Wins Reprieve: Court Vacates Order Forcing Indefinite Preservation of All ChatGPT Output Data

A U.S. federal court has issued a ruling in The New York Times’ copyright infringement lawsuit against OpenAI, vacating the directive that required OpenAI to indefinitely preserve and isolate all ChatGPT output data. According to The New York Times’ copyright infringement lawsuit against OpenAI, Judge Ona T. Wang signed the order on 9 October, bringing to an end the preservation mandate that had been in place since May 2025.

Under the court’s latest decision, effective 26 September, OpenAI is no longer obliged to preserve all ChatGPT conversation data in its entirety, though certain records must still be retained—specifically those covered by the previous order and any conversation logs linked to accounts flagged by the Times as potentially relevant. The Times may expand the scope of its designations in the future, compelling OpenAI to retain additional data for evidentiary purposes.

The ongoing lawsuit will still address whether AI models can claim fair use of copyrighted material and how journalistic works are legally positioned within generative AI.