SERVANTSTACK // INCIDENT INTELLIGENCEFORENSIC IMAGE // VERIFIED FRAMEOpenAI
A New York Times-Led News Coalition Asks a Federal Court to Sanction OpenAI, Alleging It Spent Two Years Falsely Claiming It Could Not Search Its Own Models and Logs for Their Copyrighted Journalism
On July 9, 2026, a coalition of news organizations led by The New York Times and the New York Daily News - and including the Chicago Tribune, MediaNews Group titles, Ziff Davis and the Center for Investigative Reporting - asked the federal court in Manhattan to sanction OpenAI for discovery…
- 01TRIGGEROn July 9, 2026, a coalition of news organizations led by The New York Times and the New York Daily News - and…
- 02MACHINE ACTIONAdvisory output
- 03MISSING GATERisk-based SME approval before execution
- 04IMPACTFinancial harm
The short version
On July 9, 2026, a coalition of news organizations led by The New York Times and the New York Daily News - and including the Chicago Tribune, MediaNews Group titles, Ziff Davis and the Center for Investigative Reporting - asked the federal court in Manhattan to sanction OpenAI for discovery…
Case telemetry
- INCIDENT
- SS-IR-099
- DATE
- July 9, 2026
- SYSTEM
- OpenAI
- LOCATION / SCOPE
- S.D. New York (Manhattan)
- EVIDENCE
- Alleged
- AI ROLE
- Advisory output
- HARM
- Financial harm
- SOURCES
- 2 cited records
The event
On July 9, 2026, a coalition of news organizations led by The New York Times and the New York Daily News - and including the Chicago Tribune, MediaNews Group titles, Ziff Davis and the Center for Investigative Reporting - asked the federal court in Manhattan to sanction OpenAI for discovery misconduct in their landmark copyright case, first filed in late 2023. The outlets allege OpenAI spent roughly two years telling the court it could not search its large language models and ChatGPT logs for the plaintiffs' copyrighted articles - while it had, in fact, already run such searches, even before the first newspaper filed suit. Daily News counsel Steven Lieberman told the court OpenAI had been "making misrepresentations" about that capability, choosing "obstruction" over producing the datasets and logs that would show how their journalism was used to train the model. The Times says it has already spent more than $28 million litigating the case.
What the machine did
The dispute turns on what is inside ChatGPT's underlying models. The publishers allege millions of their articles were ingested to train the system, and that the decisive evidence - whether the copyrighted text can be surfaced from the models and logs - was searchable all along. The sanctions motion is not about a single hallucinated output; it is about the provenance of the training data and OpenAI's account of its own ability to inspect the system it built. When the company that operates the model tells a court it cannot look inside it, the AI's training pipeline becomes a black box that shields the very question the litigation was meant to answer.
Where the failure landed
The coalition is seeking sanctions, including attorney fees for the effort spent recovering evidence it says was improperly withheld, in one of the most consequential AI-copyright cases in the United States. The Times has poured over $28 million into the fight, and the motion escalates the case from a dispute over fair use into a fight over candor to the court - raising the stakes for how AI developers must document and disclose what their models were trained on.
Alleged
Claims reported in litigation or public allegations; not presented here as a final finding.
SOURCE RECORD UPDATED 2026-07-09
2 cited records
- 01
- 02
Risk-based SME approval before execution
The failure pattern in this case: High-stakes output had no accountable checkpoint.
The moment the path could change
The appropriate subject-matter expert reviews the evidence, exceptions, and affected people before the output becomes action.
Autonomy is a design choice.
See the operating model that keeps AI useful while preserving human authority at consequential moments.
Compare AgenticAI and AugmentedAI →