OpenAI Forced to Indefinitely Store Deleted ChatGPT Conversations Amid NYT Copyright Lawsuit

OpenAI is facing a legal challenge that has forced the company to store deleted ChatGPT conversations indefinitely due to a court order issued as part of The New York Times' copyright lawsuit. In a recent post, Brad Lightcap, OpenAI’s Chief Operating Officer, emphasized that the company is contesting this decision, which he considers an overreach that disregards established privacy norms and undermines user protections. Last month, a court mandated OpenAI to preserve all output log data that would typically be deleted, including conversations that users request to be removed or those that should be deleted according to privacy laws. According to OpenAI's standard policy, when a user deletes a chat, it is retained for 30 days before being permanently erased. This policy is now on hold until further notice from the court. The impact of this order extends to all free, Pro, Plus, and Team ChatGPT users. However, it does not affect users of ChatGPT Enterprise or ChatGPT Edu, nor businesses with zero data retention agreements. OpenAI assured its users that the stored data will remain private, accessible only to a select, audited team within the company for legal purposes. The legal dispute began in 2023 when The New York Times filed a lawsuit against OpenAI and Microsoft, alleging that the two companies had copied and used millions of the newspaper’s articles to train their respective AI models. The publication contends that retaining user data could be crucial in preserving evidence to support its case. In response, OpenAI CEO Sam Altman posted on X (formerly Twitter), expressing strong opposition to the court's demand. “We believe this request is inappropriate and sets a dangerous precedent,” Altman said. “Protecting user privacy is one of our core principles, and we will fight any demands that compromise this commitment.” The New York Times chose not to comment on OpenAI’s statement, leaving the broader implications of the court order to be debated in the legal and tech communities. The issue highlights a growing tension between intellectual property rights and the evolving practices in AI data usage, particularly regarding user-generated content and privacy. As the case unfolds, it may have significant ramifications for how AI models are trained and how user data is managed in the future.