Microsoft exec called AI scraping the “largest theft of labor in human history”

Live updates

1. 3d ago

   Microsoft Offers Buyouts to 60,000 Workers Amid Heavy AI Investment

   Microsoft is offering voluntary severance buyouts to roughly 60,000 employees, representing just over a quarter of its total staff. The software giant is cutting headcount and trimming costs to retool its workforce while pouring billions of dollars into artificial intelligence data centers and investments. This move follows previous industry disruption from advanced models like Claude Opus 5.5, which forced video game publishers to confront automated code decompilations and unauthorized web browser ports.

   Why it matters

   Major technology companies are reshaping their operations as they redirect capital toward automated systems and artificial intelligence infrastructure. A Microsoft executive previously characterized artificial intelligence data harvesting as the largest theft of labor in human history. Meanwhile, platforms and news outlets continue blocking data scrapers while Meta prepares to enforce business account scraping terms.

   What is confirmed

   • Microsoft plans to offer buyouts to about 60,000 workers, amounting to just over a quarter of its total staff.
   • The buyout offers are part of company efforts to trim costs and retool the workforce while investing heavily in artificial intelligence and data centers.

   What to watch next

   • Monitor the acceptance rate of the voluntary buyout offers among Microsoft staff.
   • Track further announcements regarding Microsoft artificial intelligence spending and infrastructure deployment.
   Sources used for this update (4)

   1. www.ign.com — Microsoft Offer Buyouts to 60,000 Workers as It Doubles Down on AI Investment
   2. www.onmanorama.com — Is AI taking over jobs worldwide? Microsoft's buyout offer ...
   3. finance.yahoo.com — Microsoft’s Severance Offer to 60,000 Employees Is All About ...
   4. fortune.com — Microsoft offers buyouts to nearly 30% of workforce in ...
   confidence 100%
2. 6d ago

   Microsoft and Epic Games Agree AI Scraping is Theft

   Recent advances in artificial intelligence models like Claude Opus 5.5 have upended the video game industry through automated code decompilations and unauthorized web browser ports. The technology has forced publishers to confront the rapid generation of vibe-coded ports and native PC recompilations that bypass traditional development timelines. A Microsoft executive previously described artificial intelligence scraping as the largest theft of labor in human history. Technology platforms and news outlets are simultaneously moving to block data harvesting, while Meta prepares to enforce new terms allowing business account scraping without an opt-out option.

   Why it matters

   The video game industry faces a reckoning after years of publishers embracing artificial intelligence tools. Automated coding advancements now threaten proprietary source code and intellectual property at an unprecedented scale, turning developer sentiment against uncompensated web harvesting. Meanwhile, publishers continue filing regulatory complaints against aggressive data scraping practices.

   What is confirmed

   • Advancements spearheaded by Claude Opus 5.5 have transformed the coding landscape by producing rapid decompilations and unauthorized browser ports.
   • The Morning Call explicitly prohibits the use of its website content for artificial intelligence training, machine learning models, text and data mining, or algorithms without explicit written consent.

   Still unconfirmed

   • The rapid production of vibe-coded browser ports and native PC recompilations represents the video game industry experiencing karma for its prior reliance on artificial intelligence tools.
   • A Microsoft executive called artificial intelligence scraping the largest theft of labor in human history.

   What to watch next

   • Implementation of Meta new terms on 30 October allowing business account scraping without an opt-out option
   • Regulatory responses to publisher complaints regarding artificial intelligence web harvesting
   • Further industry reactions to automated code decompilations driven by Claude Opus 5.5
   Sources used for this update (3)

   1. www.mcall.com — Technology – The Morning Call
   2. kotaku.com — Microsoft And Epic Games Suddenly Agree That AI Is Theft
   3. rockstarmedia.com.my — Microsoft and Epic Suddenly Agree That AI Is Theft
   confidence 80%
3. 9d ago

   Meta New Scraping Terms Threaten Newsrooms as AI Scrutiny Grows

   Meta plans to launch new terms on 30 October allowing the scraping of any website linked to a business account without an opt-out option. Publishers are bringing complaints to regulators as artificial intelligence web harvesting intensifies. Mary Ellen Klas warned in Bloomberg that artificial intelligence threatens local newspapers, while OpenAI describes itself as existential. Meanwhile, the Australian Society of Authors reports creator concerns regarding a hybrid publisher.

   Why it matters

   Publishers and creators face mounting pressure as technology companies aggregate vast amounts of online content to train artificial intelligence models. Legal disputes and regulatory scrutiny have intensified following disclosures that OpenAI utilized secret web crawlers to copy publisher works. Industry groups continue to demand stronger copyright protections against unauthorized offshore harvesting.

   Still unconfirmed

   • Meta new terms going live on 30 October will permit scraping of any website linked to a business account with no opt-out.
   • Mary Ellen Klas warned in Bloomberg that artificial intelligence will finish off local newspapers.
   • The Australian Society of Authors has been contacted by creators about concerns regarding a hybrid publisher.

   What to watch next

   • Meta implementation of new terms on 30 October 2026
   • Regulatory responses to publisher complaints against Meta data collection
   • Further publisher legal actions regarding artificial intelligence training data
   Sources used for this update (5)

   1. www.commercial-news.com — KOLBE COLUMN: Newspapers under attack in the largest theft of ...
   2. www.blueprintforfreespeech.net — Meta ups the stakes in its game of chicken — Blueprint for ...
   3. www.asauthors.org.au — ASA calls for stronger copyright protections to address ...
   4. lindac.substack.com — The three biggest lies AI companies told us - lindac.substack.com
   5. www.ce-strategy.com — Separation | Clarke & Esposito
   confidence 80%
4. 12d ago

   OpenAI Paid Microsoft to Run Secret News Crawler

   An unsealed court filing reveals OpenAI paid Microsoft to operate a secret web crawler that copied more than 160,000 publisher works to train GPT-4o. The disclosure adds direct evidence of aggressive data collection to ongoing disputes over artificial intelligence training practices. Internal documents previously showed Microsoft staff warned about intellectual property theft, while executives debated the scale of labor displacement. Meanwhile, publishers are escalating legal actions regarding unauthorized reporting use, and technology companies continue to adjust their automated scraping parameters as regulatory scrutiny grows over data harvesting methods.

   Why it matters

   The revelation exposes the operational mechanics behind large language model training as publishers increase legal challenges over unauthorized content use. The unsealed court records directly link financial transactions to massive data collection efforts that bypassed traditional licensing agreements. These developments unfold alongside shifting industry projections regarding artificial intelligence capabilities and workforce impacts.

   What is confirmed

   • An unsealed filing shows OpenAI paid Microsoft to run a secret news crawler that copied more than 160,000 publisher works to train GPT-4o.

   What to watch next

   • Unsealing of the payment amounts and page counts associated with the OpenAI and Microsoft scraping arrangement.
   • The implementation of Meta's rewritten business terms on October 30 allowing news site harvesting with no opt-out.
   • Further publisher legal actions responding to newly revealed data ingestion practices.
   Sources used for this update (4)

   1. en.wikipedia.org — Microsoft - Wikipedia
   2. support.microsoft.com — Sign in to Microsoft 365
   3. m365.cloud.microsoft — Microsoft 365 - Sign into Copilot
   4. theinference.org — OpenAI paid Microsoft to scrape news in secret. Meta's fine ...
   confidence 100%
5. 18d ago

   Microsoft Exec Called AI Training Largest Labor Theft

   Internal Microsoft documents show executives and employees warned for years about intellectual property theft related to artificial intelligence training. One internal memo features Microsoft Applied Science director Brent Hecht labeling the ingestion of web data for artificial intelligence the largest theft of labor in human history. This revelation emerges as publishers escalate legal action over the unauthorized use of reporting, even as industry leaders moderate predictions regarding job displacement. Meanwhile, Microsoft AI CEO Mustafa Suleyman adjusted his timeline regarding workplace automation, adopting a projection that only 5% of human work will be replaced over a decade.

   Why it matters

   The disclosure of internal memos compounds legal scrutiny surrounding data scraping practices across the technology sector. Media publishers have mounted legal challenges against artificial intelligence developers for harvesting copyrighted material without permission. The Department of Justice has also intervened in copyright litigation as federal courts examine intellectual property standards.

   What is confirmed

   • Microsoft Applied Science director Brent Hecht called artificial intelligence training on web data the largest theft of labor in human history.
   • Internal documents show Microsoft executives and employees sounded alarms about intellectual property theft for years.
   • Microsoft AI CEO Mustafa Suleyman shared a prediction that only 5% of human work will be replaced by artificial intelligence in 10 years.

   What to watch next

   • Developments in federal court regarding pending publisher lawsuits and summary judgments involving copyright claims against artificial intelligence developers.
   • Further Department of Justice interventions or filings concerning artificial intelligence data scraping and intellectual property rights.
   Sources used for this update (7)

   1. www.windowslatest.com — “AI won’t take your job anytime soon”: Microsoft’s AI boss says only 5% of human work will be replaced in 10 years
   2. stocktwits.com — Will AI Take Your Job? Microsoft's AI Chief Now Points To Nobel Laureate's Research Saying Only 5% Of Hum....
   3. phemex.com — Microsoft Exec: AI Training Is Largest Theft of ... - Phemex
   4. alphathinker.app — Microsoft’s director of applied science has called the ...
   5. www.computerworld.com — Microsoft knows it robbed publishers to train AI. It just doesn’t care
   6. agihunt.info — Microsoft exec calls AI training 'largest labor… · AGI Hunt
   7. www.sundaytimes.lk — The journalists’ revolt against AI scraping | Print Edition ...
   confidence 100%
6. 23d ago

   USA Today Network Sues OpenAI Over Copyright Infringement

   The parent company of the USA Today network filed a lawsuit against OpenAI in New York federal court for allegedly utilizing reporting from 19 publications without authorization. This legal action follows unsealed court documents revealing that Microsoft Applied Science director Brent Hecht called OpenAI data ingestion the largest theft of labor in human history. Internal memos show leaders at Microsoft and OpenAI allegedly understood the legal risks associated with training artificial intelligence on unlicensed copyrighted materials. Meanwhile, the Department of Justice has intervened to support Microsoft and OpenAI as a copyright lawsuit brought by The New York Times approaches summary judgment.

   Why it matters

   Publishers face mounting pressure over artificial intelligence scraping practices as a market for inference forms with buyers setting prices. The tension between technology firms and news organizations centers on fair compensation and intellectual property rights for original journalism. Legal battles are escalating as major media networks challenge how artificial intelligence developers ingest copyrighted text for model training.

   What is confirmed

   • The parent company of the USA Today network filed a lawsuit against OpenAI in New York federal court on Thursday.
   • The lawsuit alleges that OpenAI used reporting from 19 publications, including the Courier Journal, without permission.
   • Court documents show Microsoft Applied Science director Brent Hecht called OpenAI data ingestion the largest theft of labor in human history.
   • The Justice Department intervened to support Microsoft and OpenAI as a copyright lawsuit filed by The New York Times nears a summary judgment ruling.

   Still unconfirmed

   • Internal memos indicate leaders at both Microsoft and OpenAI allegedly knew the legal risks of using unlicensed copyrighted books and news for AI training.

   What to watch next

   • A summary judgment ruling in the copyright lawsuit filed by The New York Times against Microsoft and OpenAI.
   • Progress in the New York federal court lawsuit filed by the USA Today network against OpenAI.
   Sources used for this update (4)

   1. www.savannahbusinessjournal.com — Oct. 5 – Will OpenAI and Microsoft be able to steal original ...
   2. investorshealth.com — Microsoft AI Exec: “Largest Theft of Labor in Human History ...
   3. www.wkms.org — USA Today network, including Courier Journal, sue OpenAI
   4. mediacopilot.ai — Publishers finally have AI buyers but no say in the price
   confidence 100%
7. 25d ago

   Microsoft Exec Labeled AI Scraping Labor Theft

   Court documents show Microsoft Applied Science director Brent Hecht called OpenAI data ingestion the largest theft of labor in human history. Internal memos indicate leaders at both Microsoft and OpenAI allegedly knew the legal risks of using unlicensed copyrighted books and news for AI training. Meanwhile, the Justice Department intervened to support Microsoft and OpenAI as a high-stakes copyright lawsuit filed by The New York Times nears a summary judgment ruling. This comes as ongoing industry discussions focus on AI scraping practices and data rights.

   Why it matters

   The disclosure of internal Microsoft correspondence brings fresh scrutiny to how technology companies acquire training materials for artificial intelligence systems. Legal battles involving major publishers and AI developers carry significant financial and operational consequences for the industry. Government intervention by the Justice Department signals heightened federal interest in these intellectual property disputes.

   What is confirmed

   • Microsoft Applied Science director Brent Hecht called OpenAI data ingestion the largest theft of labor in human history.
   • Leaders at Microsoft and OpenAI allegedly understood the legal risks of using unlicensed copyrighted books and news.
   • The Justice Department intervened on the side of Microsoft and OpenAI in a copyright lawsuit filed by The New York Times.

   What to watch next

   • A summary judgment ruling in the copyright lawsuit filed by The New York Times against Microsoft and OpenAI.
   Sources used for this update (2)

   1. account.microsoft.com — Microsoft account | Sign In or Create Your Account Today ...
   2. culturehive.co.uk — AI trends, news and tools with Andrew Davis. October 2026 - CultureHive
   confidence 100%
8. 27d ago

   OpenAI Memo Disclosures Add to Week 38 Tech Controversies

   The release of internal Microsoft memos describing AI training data practices as labor theft continues to drive industry attention, appearing alongside other major developments during week 38 in technology. Court documents revealed that Microsoft Applied Science director Brent Hecht called OpenAI data ingestion the largest theft of labor in human history, noting that leaders at both companies allegedly understood the legal risks of using unlicensed copyrighted books and news. The Justice Department has intervened on the side of Microsoft and OpenAI as a high-stakes copyright lawsuit filed by The New York Times approaches a summary judgment ruling.

   Why it matters

   This legal challenge threatens the foundational data collection methods used to build modern artificial intelligence models. The unsealed memos reveal internal awareness of potential copyright violations before training commenced. The Justice Department intervention adds significant weight to the defendants as the case nears a summary judgment ruling.

   What is confirmed

   • An internal January 2023 memo by Microsoft Applied Science director Brent Hecht described OpenAI training data practices as the largest theft of labor in human history.
   • Leaders at Microsoft and OpenAI allegedly understood that ingesting copyrighted books and news without licenses was legally problematic or pirated before proceeding with training.
   • The Justice Department has intervened on the side of Microsoft and OpenAI as a copyright lawsuit filed by The New York Times nears a summary judgment ruling.
   • The OpenAI hack involving a Claude-written exploit and ZCode git upload was ranked among the top tech stories of week 38 alongside the NYT versus OpenAI memos.

   What to watch next

   • The upcoming summary judgment ruling in the copyright lawsuit filed by The New York Times against Microsoft and OpenAI.
   • Further disclosures from unsealed court documents regarding artificial intelligence training data practices.
   Sources used for this update (2)

   1. arstechnica.com — Category: Policy - Ars Technica
   2. dev.to — OpenAI hack by Claude, ZCode's git upload: week 38 in tech ...
   confidence 100%
9. 29d ago

   DOJ Intervenes in NYT OpenAI Lawsuit as Scraping Memo Emerges

   The Justice Department has intervened on the side of Microsoft and OpenAI as a high-stakes copyright lawsuit filed by The New York Times nears a summary judgment ruling. The legal battle intensifies following the unsealing of court documents revealing an internal January 2023 memo by Microsoft Applied Science director Brent Hecht. Hecht described OpenAI training data practices as the largest theft of labor in human history. Leaders at both companies allegedly understood that ingesting copyrighted books and news without licenses was legally problematic or pirated before proceeding with training regardless.

   Why it matters

   The copyright lawsuits filed by The New York Times and the Authors Guild target the foundational data collection methods used by major artificial intelligence developers. OpenAI models harvest website content as part of the company business model. The ongoing litigation threatens the data ingestion practices that power commercial chatbot development.

   What is confirmed

   • The Justice Department has intervened on the side of Microsoft and OpenAI in the pending summary judgment.
   • Microsoft Applied Science director Brent Hecht wrote an internal memo in January 2023 calling training data practices the largest theft of labor in human history.
   • Leaders at Microsoft and OpenAI allegedly knew that ingesting copyrighted books and news without licenses was legally problematic.

   What to watch next

   • A summary judgment ruling in the New York Times copyright case against OpenAI and Microsoft.
   • Further legal developments regarding the Department of Justice intervention.
   Sources used for this update (4)

   1. prospect.org — Why Is Sam Altman a Free Man?
   2. journalismcentral.com — Unsealed Filing Shows Microsoft Executive Called AI Scraping ...
   3. dev.to — OpenAI lawsuit: Microsoft's 'theft of labor' memo in the NYT ...
   4. techcrunch.com — Microsoft - TechCrunch
   confidence 100%
10. 31d ago

    Microsoft Exec Called AI Scraping Largest Theft of Labor in History

    Newly unsealed court filings from September 17 show that Microsoft Applied Science director Brent Hecht wrote an internal memo in January 2023 describing OpenAI training data practices as the largest theft of labor in human history. The internal communications indicate that leaders at both companies allegedly knew that ingesting copyrighted books and news without licenses was legally problematic or pirated, but they moved forward with training regardless. These revelations stem from ongoing copyright lawsuits filed against Microsoft and OpenAI by The New York Times and the Authors Guild.

    Why it matters

    The unsealed documents offer a rare glimpse into the internal risk assessments of major technology firms as they built advanced artificial intelligence models. Legal actions from publishers and authors continue to target the data ingestion practices that underpin large language models. Meanwhile, federal regulators are increasing scrutiny on major AI laboratories.

    What is confirmed

    • Microsoft Applied Science director Brent Hecht wrote an internal memo in January 2023 calling OpenAI training data practices the largest theft of labor in human history.
    • Leadership at Microsoft and OpenAI allegedly recognized that ingesting copyrighted news and books without licenses was legally problematic or pirated before proceeding with training.
    • The documents emerged from copyright lawsuits brought by The New York Times and the Authors Guild against Microsoft and OpenAI.

    Still unconfirmed

    • The Federal Trade Commission is investigating OpenAI, Anthropic, and other leading labs.

    What to watch next

    • Additional unsealed court documents from the ongoing copyright lawsuits
    • Further regulatory actions by the Federal Trade Commission regarding leading AI labs
    Sources used for this update (5)

    1. www.commondreams.org — Welcome to the Plunderdome: How Big Tech Sold Out Its Climate Pledges to Power AI
    2. www.youtube.com — Microsoft AI Exec: “Largest Theft of Labor in Human History ...
    3. newsletter.strictlyvc.com — The FTC Is Investigating OpenAI, Anthropic, and Other Leading Labs - StrictlyVC
    4. inlevel9.com — Microsoft's Own Memo: LLMs Are Eating Their Supply Chain · Issue #284 · INLEVEL9
    5. ritholtz.com — 10 Thursday AM Reads - The Big Picture
    confidence 100%
11. 33d ago

    Microsoft Exec Called AI Scraping Largest Labor Theft

    Newly unsealed court filings from September 17 reveal that Microsoft Applied Science director Brent Hecht wrote an internal memo in January 2023 labeling OpenAI training data practices as an astonishing theft and the largest theft of labor in human history. The documents emerged from copyright lawsuits brought by The New York Times and the Authors Guild against Microsoft and OpenAI. The internal communications show leadership at both companies allegedly recognized that ingesting copyrighted news and books without licenses was legally problematic or pirated, yet they proceeded with training anyway.

    Why it matters

    The disclosures pierce the public defense of fair use that major artificial intelligence companies rely on to harvest training data. These lawsuits could fundamentally reshape how artificial intelligence firms source training materials and handle copyright protections. The revelations shift the debate from an external critique of labor exploitation to an internal acknowledgement of legal and ethical risks.

    What is confirmed

    • Microsoft Applied Science director Brent Hecht wrote an internal memo in January 2023 calling AI training on scraped news the largest theft of labor in human history.
    • Documents were unsealed on September 17 in copyright lawsuits brought against OpenAI and Microsoft.

    Still unconfirmed

    • Executives at both companies allegedly described AI training on copyrighted books as piracy and knew their actions were illegal.

    What to watch next

    • Further rulings in The New York Times and Authors Guild copyright lawsuits against Microsoft and OpenAI
    • Potential industry changes in how AI companies handle training data and licensing practices
    Sources used for this update (5)

    1. www.aimodeling.com — Unsealed filings: Microsoft exec privately called AI…
    2. www.aipulse.it — AI Copyright Case: Execs Knew Training Data Was 'Pirated ...
    3. finance.biggo.com — OpenAI Used Pirated Books for AI Training Despite Knowing ...
    4. humandriven-ai.com — AI and Labor: The 'Largest Theft of Labor' Debate — Human ...
    5. www.middleclassprepper.com — Unsealed Court Briefs Show OpenAI Executives Acknowledged ...
    confidence 100%
12. 35d ago

    Microsoft exec describes AI scraping as 'largest theft of labor in human history'

    A Microsoft executive privately described AI data scraping as the 'largest theft of labor in human history', contradicting the company's public stance that such scraping constitutes fair use. This admission was revealed through unredacted court filings in a lawsuit brought by The New York Times. The comments suggest leadership recognizes the practice as a massive appropriation of human work and a potential violation of copyright law.

    Why it matters

    The lawsuit, brought by The New York Times against OpenAI and Microsoft, has led to the unsealing of internal communications that reveal a disconnect between the company's public and private stances on AI training practices. The case has significant implications for the future of AI development and the use of online content. The revelations have sparked debate about the ethics of AI training and the need for clearer regulations.

    What is confirmed

    • A Microsoft executive described AI training as the 'largest theft of labor in human history'
    • The comments were made in internal communications revealed through unredacted court filings in a lawsuit brought by The New York Times
    • Microsoft publicly defends its AI training methods as fair use, while privately expressing concerns about the practice

    What to watch next

    • The outcome of the lawsuit brought by The New York Times against OpenAI and Microsoft
    • Further revelations about Microsoft's and OpenAI's internal communications on AI training practices
    • Regulatory developments on AI training and the use of online content
    Sources used for this update (9)

    1. www.wionews.com — Microsoft executive called AI training the ‘largest theft of labour in human history’
    2. www.searchlight.vc — The Price of Everything: A Farewell to Searchlight –A light goes out
    3. www.theringer.com — The AI Singularity Syllabus
    4. www.europesays.com — Microsoft exec: AI is a ‘doom loop’ scraping the ‘largest ...
    5. ediscoverytoday.com — The Largest Theft of Labor in Human History. Says the ...
    6. rightstech.com — Microsoft exec called AI scraping ‘the largest theft of labor ...
    7. www.wnylabortoday.com — “It’s The "Largest Theft Of Labor In Human History" - New ...
    8. fastnews.team — Microsoft exec called AI scraping ‘the largest theft of labor in human history,’ new unredacted filin....
    9. www.computeleap.com — The Emails That Could Sink Fair Use | ComputeLeap
    confidence 85%
13. 44d ago

    Microsoft Executive Calls AI Training Largest Theft of Labor in History

    Internal Microsoft communications reveal an executive described AI data scraping as the "largest theft of labor in human history." This admission contradicts the company's public legal stance that such scraping constitutes fair use. The comments surfaced through unredacted court filings in a lawsuit brought by The New York Times. While Microsoft defends its training methods in court, these private assessments suggest leadership recognizes the practice as a massive appropriation of human work and a potential violation of copyright law.

    Why it matters

    The dispute centers on whether AI companies can use copyrighted news and creative content without payment. This case highlights a gap between corporate legal strategies and internal executive views on labor ethics. It adds pressure to ongoing debates regarding AI training transparency and intellectual property.

    What is confirmed

    • A Microsoft executive labeled AI training the "largest theft of labor in human history."
    • Unredacted court filings from a lawsuit involving The New York Times revealed these internal communications.
    • Microsoft and OpenAI have publicly claimed that AI data scraping is fair use.

    What to watch next

    • Court rulings on the fair use defense in the New York Times lawsuit
    • Public responses from Microsoft regarding the leaked internal communications
    Sources used for this update (4)

    1. note.com — Claude Supports Common Instructions: New Common Sense Before Entrusting Tasks to AI
    2. note.com — 7 AI News Highlights of the Day | September 20, 2026 |｜宮野宏樹
    3. www.cryovex.com — Microsoft Exec Labels AI Training the Largest Theft of Labor in History
    4. www.martincid.com — Mustafa Suleyman: “Controlling these things are going to be a really, really big challenge for us”
    confidence 100%
14. 45d ago

    Court records reveal Microsoft executive called AI scraping "largest theft of labor"

    Unredacted court filings show Microsoft and OpenAI executives privately described AI data scraping as theft while publicly claiming fair use. One Microsoft executive specifically labeled the practice the "largest theft of labor in human history." These internal communications emerged during a lawsuit involving The New York Times. The disclosures suggest a disconnect between the companies' legal defenses and their internal assessments of how AI training impacts human labor and copyright law.

    Why it matters

    AI companies rely on massive datasets scraped from the web to train large language models. Publishers argue this destroys their business models by removing the need for users to click through to original sources. The legal battle centers on whether this constitutes fair use or copyright infringement.

    What is confirmed

    • A Microsoft executive called AI scraping the "largest theft of labor in human history"
    • Unredacted court filings show Microsoft and OpenAI executives privately called AI data scraping theft while publicly defending it as fair use
    • Internal communications were revealed through a lawsuit between The New York Times and Microsoft and OpenAI

    Still unconfirmed

    • Donald Trump supports AI scraping

    What to watch next

    • Court rulings on whether executive admissions invalidate fair use defenses
    • Further unsealing of internal OpenAI and Microsoft communications
    Sources used for this update (9)

    1. Financial Times — OpenAI staff knew the ‘existential threat’ AI posed to publishers, New York Times claims
    2. Electronic Frontier Foundation — Victory! Appeals Court Rejects Expansive New Copyright Claim
    3. arstechnica.com — Microsoft exec called AI scraping the “largest theft of labor in human history”
    4. washingtonpost.com — Microsoft exec called AI the ‘largest theft of labor’ in history, court records show
    5. Reuters — OpenAI, Microsoft executives' quotes on AI training threaten copyright defense, news outlets argue
    6. www.yahoo.com — Microsoft’s Own Executive Called AI Training “the Largest Theft of Labor in Human History”
    7. www.yahoo.com — Microsoft exec said AI scraping was ‘the largest theft of labor in human history,’ lawsuit filings reveal
    8. www.tweaktown.com — Internal Microsoft warning brands AI scraping the 'largest theft of labor in human history'
    9. hothardware.com — Microsoft Exec Warned AI Web Scraping Makes A Mockery Of Fair Use
    confidence 90%