Tech Companies Can no Longer Rely on “We Delete by Default”: OpenAI’s Recent Case
On May 13, 2025, U.S. Magistrate Judge Ona T. Wang issued a pivotal preservation order in In re: OpenAI, Inc, (S.D.N.Y., Case No. 1:23-cv-11195) compelling OpenAI to preserve all output log data that would otherwise be deleted—whether at the request of users or under international privacy laws.
This ruling, part of ongoing copyright infringement litigation brought by a coalition of media organizations (“News Plaintiffs”), carries sweeping implications beyond copyright. It directly challenges how tech companies like OpenAI reconcile user privacy rights with their legal obligations to preserve potential evidence.
What Happened?
During a discovery hearing on January 22, 2025, OpenAI revealed it was deleting certain user interactions based on its internal policies, user requests, and compliance with privacy laws like the General Data Protection Regulation (GDPR) and California Consumer Privacy Act (CCPA).
The plaintiffs raised alarm bells: critical evidence could be lost if OpenAI deletes user data that may prove infringing behavior. OpenAI resisted broad preservation demands, arguing that retaining all logs would contradict privacy law and user expectations.
Judge Wang pressed the issue, asking why OpenAI could not anonymize or segregate data rather than delete it outright. Finding OpenAI’s answers lacking, the Court ordered preservation of all output log data going forward, even if it would otherwise be destroyed under standard privacy protocols.
🔗 Read the order: Judge Wang’s May 16, 2025, Order Denying OpenAI’s Motion to Reconsider
Why It Matters—Especially for Privacy and Tech Lawyers
This isn’t just a procedural ruling. It’s a landmark flashpoint in the debate over how courts and corporations balance digital privacy, litigation obligations, and emerging technologies like generative AI.
As a cryptocurrency law firm that routinely advises on compliance and litigation involving decentralized technologies, we at Hodder Law Firm see this case as a template for what’s to come:
- Privacy laws like GDPR Art. 17 (“Right to Erasure”) and CCPA §1798.105 give users powerful rights to have their data deleted.
- But U.S. Federal Rules of Civil Procedure, especially Rule 37(e), place a duty on litigants to preserve electronically stored information that may be relevant to ongoing litigation.
What happens when these collide? This case answers: litigation trumps deletion, at least temporarily.
A Warning for the AI Industry
Tech companies can no longer rely on “we delete by default” as a defense against discovery obligations. The Court emphasized that:
“OpenAI is NOW DIRECTED to preserve and segregate all output log data that would otherwise be deleted… whether such data might be deleted at a user’s request or because of ‘numerous privacy laws and regulations.’”
— Judge Ona T. Wang, Order dated May 13, 2025
This is especially critical for developers of AI products, blockchain-based platforms, and decentralized applications. If your system handles user input—and litigation looms—you must build processes to segregate, anonymize, and preserve relevant data while honoring user trust.
Our Perspective at Hodder Law Firm
This decision confirms what we’ve long advised: privacy and litigation are not mutually exclusive. You must build a legal infrastructure that accommodates both. This ruling also signals that the judiciary is increasingly savvy about how privacy rights can be misused to sidestep accountability.
Whether you’re deploying AI to automate compliance workflows, building a virtual currency platform, or managing a P2P trading desk, the same principle applies:
- Design for compliance
- Prepare for discovery
- Don’t assume deletion is a shield
We expect more courts to adopt this posture as AI and blockchain applications become central to commercial litigation.Read the order: Wang v. OpenAI (S.D.N.Y., Case No. 1:23-cv-11195)
