AI Bots Hoover Up the Supply Chain, Microsoft Memos Warn in NYT Case 🌀

Internal Microsoft documents from 2023 described the scraping of news articles by large AI models as "the largest theft of labor in human history" and warned that the practice could trigger a "doom loop" that degraded the very models being trained, according to court filings unsealed Thursday and reported by The New York Times. The documents were disclosed in the copyright suit the Times filed against OpenAI and Microsoft in late 2023, which has since been joined by eleven other publishers. Microsoft identified the author of the memos as Brent Hecht, a director of applied science who also held a post at Northwestern University, and said in a filing that Hecht was not a decision maker and was employed to "present divergent and asymmetric perspectives," so the views expressed were not the company's. The same author wrote that large AI models "are a product that destroys its supply chain," and predicted that "millions of people around the world will soon consider large models 'hoovering up' all their work to be an astonishing theft of unprecedented proportions."

Judge Sidney Stein of the Southern District of New York is weighing summary judgment motions in the case, and a federal magistrate judge on Wednesday ordered OpenAI to produce roughly 20 million de-identified ChatGPT conversation logs to the Times and other plaintiffs under a protective framework, denying OpenAI's bid to block the production. The order deepens OpenAI's exposure across a range of copyright and data-governance disputes and could shape practices at other developers, including Anthropic. OpenAI has contested the publishers' claims throughout the litigation. Satya Nadella, Microsoft's chief executive, testified that "anything that is paywalled should be licensed by anyone who wants to use it," and said that had he known OpenAI was training on paywalled content he would have exercised Microsoft's contractual right to require it to retrain its models; a Microsoft spokesman said Nadella "spoke to broad principles" about how people find and consume information.

Internal OpenAI messages cited in the filings show staff discussing ways around publisher paywalls. A staffer told OpenAI president Greg Brockman about building a "hack" to bypass the New York Times paywall, and Brockman replied, "ah nice." Nick Turley, who led the ChatGPT team, wrote in June 2023 that AI posed an "existential threat" to publishers and in February 2024 that AI products "will get more and more substitutive as they get better," adding elsewhere that AI "products are largely substitutive, period." An OpenAI engineer wrote in February 2023 that "no matter how prominently we show the links, users won't click," a finding that cuts against the industry's frequent argument that chatbots drive traffic back to news sites. In a 2020 memo to Brockman and chief executive Sam Altman, then-policy director Jack Clark warned that the company was "creating systems that substitute for the labor of the people that define the 'culture' of society," and would "become the symbol of how Silicon Valley" treats creative work. The unsealing of further documents is expected to continue as Judge Stein weighs the pending motions.