Unsealed court documents from a high-stakes copyright lawsuit reveal that Microsoft and OpenAI executives privately worried their technology posed an "existential threat" to publishers. Microsoft's Brent Hecht called AI scraping the "largest theft of labor in human history," while internal data showed click-through rates to news sites plummeted up to 93%. Despite public "fair use" defenses, unsealed messages show OpenAI's Greg Brockman welcoming a "hack" to bypass paywalls, highlighting deep internal tensions over a "doom loop" that could destroy the AI models' own content supply chains.
Microsoft executive AI theft warnings
- ▪Microsoft Director of Applied Science Brent Hecht wrote in internal documents that scraping news for AI training by Microsoft and OpenAI was "an astonishing theft of unprecedented proportions" and perhaps the "largest theft of labor in human history."
- ▪A Microsoft spokesperson stated that Brent Hecht's comments about AI scraping of news content, made public in court filings unsealed on September 17, 2026, reflect only one employee's individual perspective, are not a legal analysis, and do not represent Microsoft's official views
OpenAI internal threat assessments
- ▪OpenAI President Greg Brockman wrote in approximately 2017 that he was "deeply motivated by the gazillions" he hoped to gain by commercializing OpenAI's technology
- ▪OpenAI President Greg Brockman responded "ah nice" when informed by researcher Nick Ryder in an internal message about a "hack" to bypass The New York Times' online paywall, as disclosed in unsealed filings from The New York Times' copyright lawsuit against OpenAI and Microsoft
- ▪OpenAI's head of ChatGPT, Nick Turley, wrote in internal messages that publishers face an "existential threat" from commercial AI products trained on news content
Copyright lawsuit unsealed documents
- ▪The New York Times and other news plaintiffs filed a motion for summary judgment, unsealed on September 17, 2026, revealing previously redacted internal emails and documents from Microsoft and OpenAI
- ▪Documents unsealed on September 17, 2026, in The New York Times' copyright lawsuit against Microsoft and OpenAI show that Microsoft provided training data to OpenAI through initiatives named Project Taxi and Project Mango, with Project Mango containing copies of at least 160,903 unique works
- ▪Filings unsealed on September 17, 2026, in The New York Times' copyright lawsuit against Microsoft and OpenAI reveal that OpenAI's mid-training datasets contained more than 91,692 copies of works published by The New York Times, Daily News, and Center for Investigative Reporting
AI substitution of journalism
- ▪Microsoft CEO Satya Nadella testified under oath that conversing with chatbots has substituted for news platforms by giving users information directly on the AI platform instead of directing them to the underlying source
- ▪Microsoft's internal data recorded click-through rate drops of 83% to 93% for some news plaintiffs, and 51% to 94% for others, when comparing its AI answer tools to traditional search
- ▪An internal Microsoft document described a "doom loop" where generative AI tools threaten the economic foundations of their essential content suppliers, ultimately hurting the performance of the AI models
Fair use defense contradictions
- ▪Microsoft CEO Satya Nadella testified that anything behind a paywall should be licensed, and that he would have required OpenAI to retrain its models had he known they scraped paywalled information
- ▪The Trump administration filed a brief on September 1, 2026, supporting OpenAI and arguing that AI training is "extraordinarily" transformative and constitutes fair use
- ▪OpenAI and Microsoft have argued in court that training AI models on news articles is protected under the "fair use" doctrine because it transforms the copyrighted material
Debatable claims
- ▪Training AI models on copyrighted news constitutes fair use
- ▪Generative AI chatbots will destroy the economic foundation of journalism
Story comments
Loading comments…