Microsoft argues Copilot rarely reproduces copyrighted text in NYT lawsuit
Source published ·Modelwire updated
Original coverage: The Verge - AI ↗·How Modelwire adds context
The development
Microsoft's legal defense in the New York Times copyright lawsuit hinges on empirical claims about Copilot's actual output behavior. Discovery filings reveal the chatbot rarely reproduces full sentences from copyrighted news or books, suggesting minimal direct substitution risk. This evidence shapes a critical inflection point in publisher litigation against generative AI systems: whether training on copyrighted material constitutes infringement if the model doesn't regurgitate substantial passages verbatim. The outcome will likely influence how courts evaluate fair use for LLM training and set precedent for similar cases against other AI labs.
Modelwire’s AI-generated summary of coverage from The Verge - AI.
Modelwire analysis
Skeptical readOur AI-generated reading of the wider context and the next developments to watch.
Microsoft's discovery filing doesn't prove Copilot poses no substitution risk; it only shows the chatbot doesn't dump full articles verbatim. The critical omission: whether partial passages, paraphrased content, or synthesized summaries still displace NYT's licensing value or reader traffic.
This filing sits in direct tension with Google's licensing pivot from early September. While Google negotiated explicit permissions with studios, Microsoft is arguing in court that training without permission is defensible because output doesn't regurgitate source material wholesale. The two strategies reflect opposite bets on whether courts will accept 'we don't copy-paste' as sufficient fair use cover. The outcome here will determine whether licensing becomes mandatory (Google's bet) or optional (Microsoft's position).
If the NYT's rebuttal discovery reveals instances where Copilot summaries closely track NYT's specific framing or reporting angles (not just facts), Microsoft's 'minimal reproduction' defense collapses. Watch whether the court schedules summary judgment motion hearings before Q1 2027; early dismissal would signal the judge finds the verbatim-output argument dispositive, while proceeding to trial suggests courts care about derivative value beyond literal copying.
This interpretation is generated from the summary above and the archive coverage cited below. Our methodology · Report an error
Coverage behind this analysis
These archive entries ground the connection in our analysis. They are ordered by source publication date, with links to our coverage and the original sources.
·The Verge - AI
Google pays studios for AI training rights instead of fighting copyright
Google is pursuing direct licensing deals with major Hollywood studios to legally train AI models on copyrighted film and television content, signaling a strategic shift away from the scraping-and-litigation model that has defined recent AI development. The move reflects growing recognition that generative AI systems require massive volumes of high-quality training data, and that studios…
MentionsMicrosoft · Copilot · The New York Times · New York Times
How this coverage is produced
Modelwire uses AI to generate summaries and context from source headlines, snippets, and selected archive coverage. Automated checks do not verify every claim, and items are not routinely reviewed by a person before publication. Zacaria Solis operates the site. Read the linked source for the full evidence and report errors through our corrections process.
Modelwire summarizes, we don’t republish. The Verge - AI originally reported this story as “Microsoft says virtually nobody was grabbing NYT articles through its chatbot”. The full content lives on theverge.com. If you’re a publisher and want a different summarization policy for your work, see our takedown page.