OpenAI, Microsoft Execs Warned AI Training Threatened Publishers

Senior executives at OpenAI and Microsoft privately acknowledged that their practice of scraping online news content to train AI models could devastate publishers, according to newly unsealed court documents in a major copyright lawsuit.

OpenAI's chief product officer Nick Turley warned internally that AI products posed an "existential threat" to news organizations, while Microsoft's director of applied science Brent Hecht described the company's approach as a "doom loop" that would harm both AI model performance and the broader web ecosystem, the Wall Street Journal reported Thursday.

Internal warnings went unheeded

The documents, made public as part of a consolidated copyright case brought by multiple news publishers against Microsoft and OpenAI, reveal a pattern of concern among technical staff about the business model's sustainability.

"It is highly unusual that an end-product threatens the economic foundations of its essential suppliers, but that is the situation we have created for our LLM business with respect to its 'content supply chain,'" Hecht wrote in internal communications.

Turley acknowledged that AI products "will get more and more substitutive as they get better," suggesting the companies understood their tools could replace rather than complement news sources. Internal OpenAI documents even referred to ChatGPT as a "modern newsstand," according to the filing.

The documents also detail discussions about circumventing publisher paywalls. In one exchange, OpenAI President Greg Brockman responded "ah nice" after a researcher found a method to bypass The New York Times paywall.

The legal battle

The New York Times initiated the lawsuit in December 2023, alleging that ChatGPT and Microsoft's Copilot used millions of its articles without permission to train AI models and answer user queries. Papers owned by Alden Global Capital, including the Chicago Tribune and New York Daily News, filed similar suits in 2024 that were later consolidated.

Publishers are seeking summary judgment before trial, citing testimony from Microsoft CEO Satya Nadella, Brockman, and Turley. Nadella testified that "anything that is paywalled should be licensed by anyone who wants to use it" for AI training, and said he would force OpenAI to retrain its models if they had used paywalled content improperly.

The Microsoft CEO also acknowledged that conversing with chatbots could substitute for visiting news websites directly.

Company responses and fair use defense

Both companies maintain their practices constitute "fair use" under copyright law, which permits limited use of copyrighted material without explicit permission for transformative purposes.

A Microsoft spokesperson told the New York Post that Hecht's comments "reflect one employee's individual perspective, are not a legal analysis, and do not represent the company's views." The company asserts its position is outlined in court filings explaining why these uses are "transformative" and consistent with copyright law.

The Justice Department recently filed a statement of interest supporting the tech companies' fair use argument. OpenAI did not respond to requests for comment.

Why it matters

This case could establish critical precedent for how AI companies access and use copyrighted content at scale. The internal documents suggest executives understood the economic harm their practices could inflict on publishers while proceeding anyway, potentially undermining fair use defenses that rely on demonstrating good faith and transformative purpose. For business leaders in media and technology, the outcome will shape both content licensing economics and the regulatory landscape for AI development.

These details were first reported by the Wall Street Journal based on the newly unsealed court filings.