AI Regulation Tracker / Litigation filing

News Publishers File Sanctions Motion Accusing OpenAI of Spoliation in SDNY Copyright Fight

Regulatory summary: On July 9, 2026, five news organizations led by The New York Times filed a motion in the Southern District of New York asking the court to sanction OpenAI. The motion alleges spoliation and lack of candor, including a roughly two-year false claim that OpenAI could not search its systems.

The New York Times and four other news organizations have asked a federal judge to sanction OpenAI, alleging it falsely claimed for roughly two years that it could not search its models for their work and that it deleted or hid billions of ChatGPT conversations. These are allegations in a party motion. No court has ruled.

Primary source

By Anthony Guerriero, Founder, The Leveraged Years · Reviewed by The Leveraged Years Editorial Desk · Published July 9, 2026 · Last updated July 9, 2026

News Publishers File Sanctions Motion Accusing OpenAI of Spoliation in SDNY Copyright Fight regulation briefing

The Leveraged Years AI Regulation Tracker

Key takeaways

Plaintiffs escalated from a discovery dispute to a formal request for sanctions, alleging spoliation of evidence (deleted or unsearchable ChatGPT conversations) and lack of candor to the tribunal (a claimed inability to search that plaintiffs allege was false).
Litigators and e-discovery teams handling AI matters, general counsel and in-house counsel at AI developers and enterprises deploying AI, records-management professionals, and AI vendors whose logging and retention practices could become discovery targets.
Status: Motion filed July 9, 2026.
If you run or advise on AI systems, pull your current litigation-hold and retention policies for model training corpora, output logs, and conversation data, and confirm you can defensibly represent what is and is not searchable before you ever make that representation to a court or regulator.

Regulatory Briefing

Bindingness: Litigation filing / party motion; no court finding. This is a request for relief filed by plaintiffs. It carries no legal force unless and until a judge grants it.
Scope: In re OpenAI copyright litigation (the consolidated matter that includes NYT v. Microsoft and OpenAI), U.S. District Court for the Southern District of New York. Moving parties are The New York Times, New York Daily News, the Center for Investigative Reporting, The Intercept, and Ziff Davis.
What Changed: Plaintiffs escalated from a discovery dispute to a formal request for sanctions, alleging spoliation of evidence (deleted or unsearchable ChatGPT conversations) and lack of candor to the tribunal (a claimed inability to search that plaintiffs allege was false).
Who This Affects: Litigators and e-discovery teams handling AI matters, general counsel and in-house counsel at AI developers and enterprises deploying AI, records-management professionals, and AI vendors whose logging and retention practices could become discovery targets.
Current Status: Motion filed July 9, 2026. Not yet ruled on. OpenAI has not responded on the merits in the record reviewed here and has not been found to have engaged in the alleged conduct.
Effective Date: Not applicable. This is a pending motion, not a rule, order, or statute.
Primary Source: The SDNY docket in the consolidated OpenAI copyright litigation. The filing itself is PACER-gated; obtain the docket entry and the motion through CourtListener/RECAP or PACER before relying on any specific quotation, page, or paragraph.
Secondary Analysis: Reuters (carried by U.S. News, AOL, and Yahoo Finance), Variety, Bloomberg Law, and The Associated Press (carried by The Columbian and the Arkansas Democrat-Gazette) reported the filing on July 9, 2026.
Workflow Impact: Sharpens standards for litigation holds over AI systems, e-discovery of models and training data, and candor to the tribunal when a party represents that data cannot be searched or produced.
Operational Next Step: If you run or advise on AI systems, pull your current litigation-hold and retention policies for model training corpora, output logs, and conversation data, and confirm you can defensibly represent what is and is not searchable before you ever make that representation to a court or regulator.
Last Verified: July 9, 2026

Date	Jurisdiction	Rule	Affected professionals	Status or effective date
2026-07-09	United States	Plaintiffs escalated from a discovery dispute to a formal request for sanctions, alleging spoliation of evidence (deleted or unsearchable ChatGPT conversations) and lack of candor to the tribunal (a claimed inability to search that plaintiffs allege was false).	Litigators and e-discovery teams handling AI matters, general counsel and in-house counsel at AI developers and enterprises deploying AI, records-management professionals, and AI vendors whose logging and retention practices could become discovery targets.	Motion filed July 9, 2026. Not yet ruled on. OpenAI has not responded on the merits in the record reviewed here and has not been found to have engaged in the alleged conduct.

Frequently Asked Questions

Has a court found that OpenAI deleted evidence or lied?

No. As of July 9, 2026, this is a motion filed by the news publishers. It contains allegations. No judge has ruled on it, and OpenAI has not been found to have engaged in the alleged conduct.

Who filed the motion and where?

The New York Times, New York Daily News, the Center for Investigative Reporting, The Intercept, and Ziff Davis filed it in the U.S. District Court for the Southern District of New York, in the consolidated OpenAI copyright litigation.

What sanctions are the publishers asking for?

According to press accounts, they seek sanctions including attorneys' fees and a court finding that OpenAI's chat logs showed it misused their copyrighted works. Granting any of this is entirely the judge's decision.

Can I read the actual motion?

The filing is on the SDNY docket via PACER and is access-gated. We could not retrieve a public copy. Obtain it through CourtListener/RECAP or PACER before relying on specific quotations, paragraphs, or page cites.

What should my legal or records team do now?

Review your litigation-hold and retention policies for AI training data, output logs, and conversation stores. Confirm you can accurately state what is and is not searchable before making any such representation to a court or regulator.