AI training built on fair use looks shaky when the companies' own people call it "astonishing theft"
Source: The Decoder (opens in a new tab) · Matthias Bastian
Intel Summary
Internal emails and sworn testimony have surfaced challenging OpenAI and Microsoft's fair use defense in AI training. According to reported records, a Microsoft director termed the practice the "largest theft of labor in human history," while OpenAI's head of ChatGPT noted their products are "largely substitutive, period."
Why It Matters
Admissions of market substitution and labor appropriation directly challenge the legal argument that generative AI training constitutes transformative fair use. If credited in litigation, these internal statements could expose model developers to substantial copyright liabilities and force mandatory data licensing models.
Part of an ongoing development
Independent reportingUnsealed court filings reveal Microsoft executive calling AI scraping theft
Newly unsealed court filings reveal that a Microsoft executive privately characterized OpenAI's data harvesting as "theft" and "the largest theft of labor in human history." The records indicate that both companies scraped paywalled content from The New York Times to build datasets, despite internal warnings that doing so would financially undermine publishers. Claims are as reported; this summary makes no determination about accuracy or significance.
- Confidence
- High confidence
- Corroboration
- Corroborated
More coverage of this development
- Microsoft exec called AI scraping ‘the largest theft of labor in human history,’ new unredacted filings revealTechCrunchIndependent reporting
Organizations & Entities
Related Intelligence
- ReportAlso involving ChatGPT
OpenAI's upcoming chat-and-work merger in ChatGPT signals the end of all-you-can-prompt pricing
OpenAI plans to merge Chat and Work modes into a unified interface starting in December, according to The Decoder. Under the change, Plus, Pro, and Business subscriptions will share a token-based quota across ChatGPT and Codex, where complex models consume more allocation while simple instant chats remain uncounted.
The Decoder - ReportAlso involving ChatGPT
Exclusive: OpenAI Sign Deals With Magnite, PubMatic to Fill Unused ChatGPT Ad Space
OpenAI has signed agreements with ad tech companies Magnite and PubMatic to sell unused advertising inventory on ChatGPT, according to reporting by The Information citing people familiar with the deals. The platforms' software is being used by advertisers to purchase placement.
The Information - ReportAlso involving ChatGPT
OpenAI slashes funding target by $10bn as tech anxieties mount
Financial Times reports that OpenAI has reduced its fundraising target by $10 billion, now seeking approximately $20 billion at a $1.25 trillion valuation amid growing broader tech market anxieties.
Financial Times (AI) - ReportAlso involving ChatGPT
OpenAI is steering ChatGPT users toward more paid usage
OpenAI has implemented limits on the number of paid ChatGPT accounts permitted per device and payment card. Under the new policy, users who reach the limit cannot initiate new subscriptions or access affected accounts on that device, with logging out failing to release the assigned slot. The measure restricts users attempting to bypass usage limits through multiple lower-tier subscriptions.
The Decoder