This week, OpenAI announced that free users will gain access to the ChatGPT Memory feature, allowing the AI to remember past conversations and improve future interactions. However, a recent court ruling now requires OpenAI to retain all user chats indefinitely, including those users have deleted.
The order arises from lawsuits filed against OpenAI by several news organizations, including the New York Times. (Disclosure: Ziff Davis, parent company of Mashable, also filed a lawsuit in April claiming copyright infringement related to OpenAI’s training data.)
On May 13, U.S. Magistrate Judge Ona T. Wang in New York directed OpenAI to “preserve and segregate all output log data that would otherwise be deleted,” until further notice. This mandate includes chat logs that OpenAI might normally delete.
OpenAI Challenges the Ruling
Although the ruling was issued weeks ago, it has recently become public as OpenAI seeks to block the order through oral arguments. OpenAI argues that maintaining all deleted chats would impose a “substantial burden” and require major changes to its data infrastructure. Moreover, preserving deleted data could conflict with the company’s privacy policies.
Plaintiffs argue that preserving these logs is necessary to prevent OpenAI from deleting potentially incriminating evidence, such as users evading paywalls by having ChatGPT summarize restricted articles. OpenAI counters that such claims are speculative.
If upheld, the ruling means millions of ChatGPT users should assume that every conversation — including those deleted — is permanently stored, raising significant privacy questions.
Author’s Opinion
This ruling highlights the tension between legal accountability and user privacy in the AI era. While it is crucial to ensure companies do not misuse data or evade legal scrutiny, forcing indefinite storage of all conversations could undermine user trust. Clear, balanced policies are needed that protect individual privacy rights without compromising transparency and lawful investigation.
