Technology
Microsoft argues in court filings that Copilot rarely duplicates copyrighted news articles
Defending against copyright claims from The New York Times and authors, Microsoft cited chat log discovery data to argue its chatbot seldom outputs substantive text from original articles.
The short version
- Microsoft submitted new legal filings contesting copyright infringement claims brought by The New York Times, other publishers, and book authors.
- The company provided 8.2 million Copilot chat logs filtered for publisher-related keywords to an expert hired by the plaintiffs.
- Microsoft asserts the logs demonstrate that Copilot rarely reproduces full sentences or meaningful excerpts of protected works.
- The outcome of the copyright litigation remains pending before the court.
Key facts
- Microsoft filed legal documents defending against copyright infringement lawsuits brought by publishers including The New York Times and various book authors.[The Verge]
- During discovery, Microsoft handed over 8.2 million Copilot chat logs to an expert representing the plaintiffs.[The Verge]
- Microsoft stated that the supplied logs were selected because they contained keywords referencing the plaintiff publishers' websites.[The Verge]
- Microsoft claims the data shows Copilot rarely outputs full sentences or substantive excerpts that could substitute for original copyrighted news articles or books.[The Verge]
What remains uncertain
- Whether the court or plaintiffs' analysis will validate Microsoft's characterization of the 8.2 million Copilot logs remains undetermined.[The Verge]
Sources
Outlet counts describe coverage, not independent confirmation. Reports may share a wire service or original source.