The New York Times’ copyright case against OpenAI and Microsoft has reached a major procedural milestone, with the parties moving beyond summary-judgment briefing and closer to a ruling on whether artificial-intelligence companies can use copyrighted journalism to develop commercial systems.
The dispute is part of consolidated copyright litigation in the U.S. District Court for the Southern District of New York.
The case centres on allegations that OpenAI and Microsoft used protected news content in the development and operation of AI products without permission or payment.
Table of Contents
What changed in the New York Times case
The court’s schedule required the parties to complete the next round of summary-judgment filings in October 2026.
Public docket records show that the New York Times, Daily News and Ziff Davis were among the news plaintiffs filing materials in support of their claims, while the defendants have argued that AI training is protected by fair-use principles.
Summary judgment allows a judge to resolve legal issues without a full trial when there is no genuine dispute over facts that would require a jury’s determination.
In this case, however, the volume of evidence and the competing technical and market arguments mean the court could still leave some issues for later proceedings.
As of October 17, 2026, no final decision had been announced on the central fair-use questions.
The case is therefore entering a period in which the judge may rule on some claims, narrow the issues or determine that a trial is necessary.
Why the copyright fight matters
The New York Times argues that its journalism was copied and used to build systems that can generate answers resembling or substituting for original reporting.
The company has sought damages and other remedies, while also challenging the use of its articles as training material.
OpenAI has disputed the newspaper’s interpretation of the case and says the court has narrowed some of the claims as the litigation has progressed, according to the company’s legal response.
The company maintains that AI systems learn statistical relationships from training material rather than simply storing and republishing entire works.
Microsoft, which works closely with OpenAI and incorporates AI tools into products including Copilot, is also a defendant.
Its potential liability depends on questions including what material was used, how the systems were developed and whether the company contributed to alleged infringement.
The legal arguments extend beyond one newspaper.
The consolidated proceeding includes other authors and publishers, making the eventual rulings important for licensing negotiations, online journalism and the economics of generative AI.
The main legal questions before the court
- Was protected material copied? The plaintiffs argue that copyrighted articles were reproduced or ingested during the development of AI models.
- Does fair use apply? OpenAI and Microsoft argue that model training is transformative and does not provide the original works as direct substitutes.
- Did the systems harm existing markets? The publishers say AI-generated summaries and answers can compete with subscription journalism and reduce traffic to original reporting.
- What remedy would be appropriate? Potential remedies could include damages, limits on the use of protected works or other measures, although the court has not decided what relief is available.
The United States government has urged a broad approach to fair use in AI-training disputes, while acknowledging that the analysis is fact-specific rather than an automatic rule for every model or dataset.
What happens next
The next significant stage is the court’s review of the competing motions and supporting evidence.
The parties’ reply briefing is scheduled for November 2026 under the current case timetable, after which the judge could issue a decision or set additional proceedings.
A ruling in favour of the publishers could strengthen demands for licensing agreements and compensation when AI developers use journalism in training or retrieval systems.
A ruling favouring the defendants could make it harder for publishers to challenge model training under existing U.S. copyright law.
Any U.S. decision would not automatically determine how Canadian courts interpret the Copyright Act.
Canadian publishers, technology companies and creators could nevertheless watch the case closely because it may influence contract negotiations, product design and future litigation on both sides of the border.
The case remains unresolved, and the court’s eventual decision may address only some of the claims.
The broader legal debate over who controls news content used by AI systems is expected to continue regardless of the outcome.
Frequently Asked Questions
What is the New York Times suing OpenAI and Microsoft over?
The New York Times alleges that copyrighted journalism was used without permission to develop and operate AI systems and related products.
Has the New York Times won the AI copyright case?
No. As of October 17, 2026, the court had not issued a final ruling on the central claims.
What is summary judgment?
Summary judgment allows a judge to decide issues without a full trial when the relevant facts are not genuinely disputed.
Why is fair use important in the case?
OpenAI and Microsoft argue that AI training is a transformative use, while the publishers say copying their work and creating competing products requires permission or payment.
Could the case affect Canada?
A U.S. ruling would not directly decide Canadian law, but it could influence international licensing discussions, AI product policies and future copyright disputes.
Fact-Checked: Key procedural details and the case status were checked against federal court docket information and an official OpenAI legal-response page.
Disclaimer: This article describes an active U.S. legal proceeding and does not predict the court’s final decision.