Unredacted NYT filings show Microsoft exec called AI scraping ‘largest theft of labor,’ detail paywall bypassing

Topic: technologyRegion: north americaUpdated: i2 outletsSources: 5Spectrum: Center Only⏱ 2 min read📡 Wire pickup

📰 Scored from 2 outletsacross 2 Center How we score bias →

Story Summary

SITUATION

Unredacted New York Times court filings show a Microsoft executive privately called AI scraping “the largest theft of labor in human history,” and the filings detail paywall bypassing, mass scraping and stripping copyright notices (per techcrunch). The filings also quote OpenAI leaders warning their models posed an “existential threat” to publishers and journalists, a framing the Times uses to challenge OpenAI’s fair-use defense (per techcrunch).

Coveragetap to expand ▾

Spectrum: Center Only🌍US: 1 · Other: 1

Political Spectrum

Position is inferred from coverage mix.

i2 outlets · Center

Left

Center

Right

Left: 0

Center: 2

Right: 0

Geography Coverage

Distribution of where coverage is coming from.

i2 unique outlets · Dominant: US/Canada

All2 US/CA1 · 50%Global1 · 50%

KEY FACTS

OpenAI leaders are quoted in the filings as warning that their models posed an “existential threat” to publishers and journalists (per techcrunch)
The disclosures in the unredacted filings are presented by The New York Times to sharpen its claims and to undercut OpenAI’s fair-use defense (per techcrunch)
TechCrunch reports that the newly revealed material was filed in The New York Times’ copyright suit against OpenAI and Microsoft (per techcrunch)

HISTORICAL CONTEXT

An active legal and regulatory clash over whether large AI models may lawfully ingest and reproduce copyrighted news and journalism has framed recent debates between publishers and tech firms throughout 2023–2026.

The structural rules at issue derive from the Copyright Act of 1976 (17 U.S.C.), the fair-use doctrine codified in 17 U.S.C. §107, and the Digital Millennium Copyright Act of 1998, plus pivotal court precedents such as Campbell v. Acuff‑Rose Music, Inc. (U.S. Supreme Court, March 8, 1994) and the Authors Guild v.

Brief

Unredacted excerpts filed in The New York Times’ copyright suit against OpenAI and Microsoft show executives at the companies privately describing large-scale AI scraping of news content in stark terms.

The newly revealed material includes a Microsoft executive calling the scraping effort “the largest theft of labor in human history,” and it quotes OpenAI leaders warning that their models posed an “existential threat” to publishers and journalists, details TechCrunch extracted from the filings (per techcrunch).

The filings describe specific practices plaintiffs say undercut news revenue: systematic paywall bypassing, mass scraping of articles and removal of copyright notices from scraped content (per techcrunch).

The Times frames those facts as direct evidence that OpenAI and Microsoft relied on copyrighted journalism to train models and to displace publishers’ value — a legal strategy meant to weaken OpenAI’s fair-use defense (per techcrunch).

OpenAI and Microsoft have defended their work in other public statements, but the unredacted filings shift the factual record toward internal acknowledgments reported by The New York Times and summarized by TechCrunch (per techcrunch).

Why now: the suit has moved into phases where previously redacted internal communications are being disclosed to the court, and those communications now provide contemporaneous characterizations of how company leaders viewed massive news-data ingestion (per techcrunch).

The filings do not quantify how many articles were scraped or give detailed timelines of specific scraping operations in the TechCrunch excerpt, so the court will likely test whether these internal descriptions meet the legal thresholds for copying and market harm asserted by The New York Times (per techcrunch).

Why it matters

Newsroom reporters and The New York Times stand to bear concrete revenue and labor costs if the court accepts the Times’ claim that paywall bypassing and mass scraping removed sources of subscription income (per techcrunch).
Publishers and journalists face reputational and economic risk because OpenAI leaders described the models as posing an “existential threat” to their business in internal communications (per techcrunch).
Microsoft and OpenAI could incur major legal and financial liabilities if courts treat their internal acknowledgment of large-scale scraping as evidence against a fair-use defense (per techcrunch).

What to watch next

Whether The New York Times files additional unredacted internal communications from OpenAI or Microsoft ahead of upcoming hearings in the copyright case (per techcrunch).
Whether a court rules that the disclosed internal descriptions — including the Microsoft executive’s “largest theft of labor in human history” line — are admissible evidence against OpenAI’s fair-use defense (per techcrunch).
Whether OpenAI or Microsoft respond with counter-evidence quantifying datasets, filtering steps, or licensing efforts by the next major filing deadline in the case (per techcrunch).

Where sources differ

7 dimensions

Framing differences

Only TechCrunch is in this pack; it frames the unredacted filings as strengthening The New York Times’ suit and highlighting internal alarm from Microsoft and OpenAI (per techcrunch)

Disputed or unclear

TechCrunch reports the filings describe paywall bypassing and mass scraping but does not provide quantitative figures or timelines for those practices (per techcrunch)

Omitted context

No source text in this pack quantifies how many articles were scraped or the exact period during which paywall bypassing occurred; readers lack dataset sizes and timelines (per techcrunch)
No source text in this pack names specific Microsoft or OpenAI executives by verbatim name in the excerpt provided here, leaving gaps on which individuals made the quoted statements (per techcrunch)
No source text in this pack cites licensing offers or negotiations between publishers and AI companies that would clarify whether commercial options existed before scraping (per techcrunch)
No source text in this pack references any prior court rulings on AI training data that directly govern this case (per techcrunch)

Conflicting figures

TechCrunch does not provide numerical estimates for articles scraped, user counts affected, or dollars of alleged market harm (per techcrunch)

Disputed causality

TechCrunch reports internal quotes describing scraping and existential risk but does not establish a direct legal causality link between those actions and measured publisher revenue losses (per techcrunch)

Attribution disputes

TechCrunch attributes the quotes and descriptions to unredacted material in The New York Times’ copyright suit; it does not attribute the quotes to named individuals in the excerpt provided (per techcrunch)

Related Developments1 story

Microsoft and OpenAI warned their AI would 'hurt the web' and threaten news business, court filing shows

OpenAI and Microsoft executives told a court their AI models ingest news articles and could create a ‘doom loop’ that threatens the commercial foundations of news publications (per Washington Examiner). The revelation appears in a largely unredacted filing unsealed September 17, 2026 as part of The New York Times' copyright suit (per Washington Examiner).

1d ago

›

Sources

5 of 5 linked articles