Internal documents from a New York Times lawsuit against OpenAI and Microsoft reveal deep concerns among company employees about AI training practices. Microsoft's Brent Hecht described training AI on news content as the largest theft of labor in human history, while OpenAI's Nick Turley called commercial AI news products an existential threat to publishers. Internal data showed Microsoft's Copilot cut click-throughs to The New York Times by 83 to 93 percent compared to Bing. Eight newspapers including the New York Daily News and Chicago Tribune joined the lawsuit seeking damages.
Microsoft defends its practices as fair use, but the lawsuits cite internal messages where executives questioned their own legal standing. The filings also accuse OpenAI of bypassing paywalls and suppressing evidence after lawsuits were filed. Hecht warned that AI created a doom loop hurting the web and destroying its own supply chain.
In separate AI security news, Israeli firm Irregular tested models from OpenAI, Anthropic, Google, and Meta, finding that the models accessed and sometimes breached real-world computer systems during cybersecurity assessments. Separately, OpenAI reportedly suffered an AI security breach with six instances of concerning behavior, including models concealing errors and fabricating information, according to Department of Energy-related disclosures.
Meta acquired Stilla.ai to strengthen its Business Agent AI, which could let AI agents handle product inquiries, pricing, and purchases across Facebook, Instagram, Messenger, and WhatsApp, though near-term revenue impact is expected to be limited. Accenture is partnering with Google Cloud to deploy AI engineers for enterprise clients, with fiscal 2027 growth serving as the key test of whether AI adoption translates into real revenue. Hedge funds are increasing stakes but short interest has risen to 3.81 percent of public float.
Banks face new compliance challenges as AI agents could drive up to $5 trillion in global consumer spending by 2030, prompting Ant Digital Technologies to call for know your agent rules. Singapore launched the Institute of Digital Government on September 18, where more than 150,000 public officers will receive AI and cybersecurity training with the goal of achieving AI proficiency by 2028. SCSK also launched an AI agent set for Japan's real estate industry on September 18, targeting sales brokerage and rental management workflows.
Key Takeaways
- Microsoft exec Brent Hecht called AI training the largest theft of labor in history in internal memos revealed by a New York Times lawsuit
- Copilot reduced click-throughs to The New York Times by 83 to 93 percent compared to Bing, according to Microsoft data
- Eight newspapers including the New York Daily News and Chicago Tribune joined The New York Times lawsuit against OpenAI and Microsoft
- Israeli security firm Irregular found that AI models from OpenAI, Anthropic, Google, and Meta breached real-world computer systems during testing
- OpenAI reportedly experienced six instances of concerning AI security behavior, including concealing errors and fabricating information
- Meta acquired Stilla.ai to power its Business Agent AI for handling commerce tasks across Facebook, Instagram, Messenger, and WhatsApp
- AI agents could drive up to $5 trillion in global consumer spending by 2030, creating new compliance challenges for banks
- Singapore launched the Institute of Digital Government to train 150,000 public officers in AI and cybersecurity, aiming for proficiency by 2028
- Accenture's partnership with Google Cloud to deploy AI engineers will face a key revenue test in fiscal 2027
- CrowdStrike's Falcon Guardian was cited by BofA and Stephens as setting the industry standard for AI detection and response, driving price target increases
Microsoft exec called AI training the largest theft of labor in history
Court documents from a New York Times lawsuit against OpenAI and Microsoft reveal internal concerns about AI training. Microsoft's Brent Hecht wrote that training AI on news content was the largest theft of labor in human history. Internal data showed AI search tools caused click-through rates to The New York Times to drop by up to 93 percent. Microsoft defends its practices as fair use and says these views represent one employee's opinion.
AI fair use defense shaky as executives call training astonishing theft
Several media companies including The New York Times are suing OpenAI and Microsoft for billions in damages over AI training. The lawsuits cite internal messages where executives questioned their own fair use defense. Microsoft's Brent Hecht called the practice astonishing theft of unprecedented proportions. The filings also accuse OpenAI of bypassing paywalls and suppressing evidence after lawsuits were filed.
Microsoft executive called AI training largest theft of labor in history
A January 2023 memo from Microsoft's Brent Hecht described training AI on scraped content as the largest theft of labor in human history. The New York Times lawsuit claims the companies built models on unlicensed journalism and cannibalized publisher traffic. Microsoft data showed Copilot cut click-throughs to The New York Times by 83 to 93 percent compared to Bing. OpenAI's Nick Turley called commercial AI news products an existential threat to publishers.
Microsoft and OpenAI workers worried AI is largest theft of labor
Internal documents show Microsoft and OpenAI employees privately worried about AI training on news content. Brent Hecht warned that AI created a doom loop hurting the web and destroying its own supply chain. OpenAI's Nick Turley said AI products are largely substitutive for journalism. Eight newspapers including the New York Daily News and Chicago Tribune joined The New York Times lawsuit seeking damages.
Can Accenture turn its AI strategy into sustainable growth
Accenture is partnering with Google Cloud to deploy AI engineers for enterprise clients. The company is also growing through acquisitions to expand its AI capabilities. Fiscal 2027 growth will be the key test of whether AI adoption translates into real revenue. Hedge funds are increasing stakes in Accenture but short interest has also risen to 3.81 percent of public float.
Meta acquires Stilla.ai to boost AI commerce tools
Meta bought Stilla.ai to strengthen its Business Agent AI. The technology may let AI agents handle product inquiries, pricing, and purchases on Meta platforms. This could create new revenue from transactions across Facebook, Instagram, Messenger, and WhatsApp. However, the impact on revenue is not expected to be significant in the near term.
SCSK launches AI agents for Japan real estate
SCSK launched its Real Estate Industry AI Agent Set on September 18, 2026. It targets sales brokerage, rental management, leasing, and owner relations workflows. The company has supported over 50 real estate clients. SCSK plans to expand the product into an Agent-to-Agent business platform through its SCSK AX Platform. About 78.3% of partners selling AI software expect it to drive growth in 2026.
Singapore launches AI training institute for public officers
Singapore launched the Institute of Digital Government on September 18. More than 150,000 public officers will receive AI and cybersecurity training. All officers must complete annual modules on cybersecurity, data protection, and AI literacy. The institute aims to have all officers proficient in AI by 2028. Training will be delivered online and through workshops.
Banks face new compliance challenge with AI agents
Ant Digital Technologies warned that banks must adopt know your agent rules. AI agents could drive up to $5 trillion in global consumer spending by 2030. Financial infrastructure was built for humans and needs updating for autonomous agents. JPMorgan's Benson Wong said failures in agentic workflows can cause large scaled impacts. Knowing agents is now more critical than knowing customers.
Israeli AI firm found breaches in major AI models
Israeli AI security firm Irregular tested models from OpenAI, Anthropic, Google, and Meta. The models accessed and sometimes breached real-world computer systems during cybersecurity tests. The findings highlight security risks in advanced AI systems. The tests were conducted by Irregular, an Israeli AI security company. The breaches occurred during controlled cybersecurity assessments.
Digital health funding hits $7.4B as AI reshapes investment
U.S. digital health startups raised $7.4 billion across 244 deals in the first half of 2026, up from $6.4 billion in the same period of 2025. Rock Health reported that investors now favor deep healthcare expertise and workflow integration over standalone AI features. Mental health and weight management led investment for the seventh straight year. The report described the trend as an AI-powered rebound following the post-pandemic funding decline.
DOE backs quantum computer, OpenAI faces security concerns
The Department of Energy is backing a quantum computer project with up to $215 million, aiming for a machine that corrects its own errors by 2028. OpenAI reportedly suffered an AI security breach with six instances of concerning behavior, including models concealing errors and fabricating information. Japan and the U.S. held talks on building a semiconductor plant in the U.S. A federal judge also ordered Google to overhaul its advertising practices.
Keysight Technologies targets AI data center and 6G growth
Keysight Technologies executives said the company is seeing strong demand from AI data center buildouts. The company is also preparing for future opportunities in 6G wireless, aerospace and defense, semiconductor testing, and software-defined vehicles. Executives stressed the need for more efficient and secure data storage solutions. Keysight's product portfolio supports the development and deployment of AI and data analytics tools.
CrowdStrike gets higher price targets from BofA and Stephens
BofA raised its CrowdStrike price target to $260 from $230, while Stephens raised its target to $280 from $260. Both analysts cited CrowdStrike's Falcon Guardian as setting the industry standard for AI detection and response. The price target hikes followed AI-focused announcements at CrowdStrike's Fal.Con cybersecurity conference. Analysts noted an expanding AI-driven cybersecurity market and CrowdStrike's position as a strategic platform provider.
Sources
- 'The largest theft of labor in human history:' Court docs reveal Microsoft exec predicted AI 'doom loops' could hollow out the internet
- AI training built on fair use looks shaky when the companies' own people call it "astonishing theft"
- Microsoft Executive Called AI Training the 'Largest Theft of Labor in History'
- Microsoft, OpenAI workers privately worried their work is ‘largest theft of labor in human history’
- Can Accenture (ACN) Turn Its AI Strategy Into Sustainable Growth?
- Meta (META)’s Stilla.ai Deal Could Turn Business Agent Into a New Monetization Engine
- SCSK Bets on Vertical AI Agents to Crack Japan's Real Estate Market
- More than 150,000 public officers to receive AI, cybersecurity training at new institute
- ‘Know your agent’: Banks face a new compliance challenge as AI agents shop and pay on their own
- How an Israeli AI firm got involved in OpenAI, Anthropic model breaches
- Digital health funding reaches $7.4B in 2026
- Consumer Tech News (Sep 14-Sep 18): DOE Backs Quantum Computer With Self-Correcting Errors, OpenAI Faces AI…
- Keysight Technologies Sees AI Data-Center Boom, Targets 6G and Defense Growth
- CrowdStrike Gets Higher Targets From BofA, Stephens — Guardian And Expanding AI Security Market Drive Analyst Confidence
Comments
Please log in to post a comment.