Newly unsealed court filings in The New York Times' copyright lawsuit against OpenAI and Microsoft reveal the scale of news content scraping used for AI training. The documents show that OpenAI's datasets contained over 91,692 copies of works from The Times, The Daily News, and the Center for Investigative Reporting. Microsoft's Copilot search engine reportedly cut traffic to The Times' website by up to 93 percent.
Inside the companies, executives privately acknowledged the severity of the issue. Microsoft's Brent Hecht reportedly called the scraping the largest theft of labor in human history, while Microsoft CEO Satya Nadella testified that paywalled content should be licensed before use in training. OpenAI's ChatGPT head Nick Turley described the product as increasingly substitutive for publisher content.
Separately, OpenAI published proposals calling for international cooperation on frontier AI safety standards, with particular focus on recursive self-improvement, or RSI. The company warned that fully autonomous RSI could lead to humans losing practical control over AI development and should not be pursued unless done safely.
In the financial sector, the Financial Conduct Authority is exploring a new regulatory framework that would treat agentic AI the same as human financial advisers. Some experts caution that this approach could deprive millions of investors of valuable AI-driven insights, while supporters argue AI can offer recommendations that human advisers cannot.
AI agents are increasingly making global supply chain decisions without human approval, from rerouting ships to reallocating inventory. IDC predicts 45 percent of G2000 companies will adopt agentic AI supply chain orchestration by 2029, with procurement seeing the biggest gains due to the measurable financial impact of each decision.
Nvidia reported $96.22 billion in revenue driven by AI infrastructure demand and expects 70 percent growth next fiscal year, with a market cap of $5.37 trillion. The company's long-term growth prospects are seen as significantly stronger than Eli Lilly's $22.97 billion in GLP-1 drug sales, though Eli Lilly holds a $1.03 trillion market cap appealing to dividend-focused investors.
On the policy front, Republican candidate Monique DeSpain and Democrat incumbent Val Hoyle are clashing over AI and data centers in Oregon's 4th Congressional District. DeSpain cited $292,000 in independent expenditures from a pro-AI PAC during the primary, while Hoyle denied accepting money from AI companies and said she supports federal regulation.
Rapid City officials warned residents to verify information after an AI-generated search result falsely claimed the city reimburses property owners for sidewalk repairs. The fabricated claim blended Rapid City rules with policies from Ohio and Indiana. A study found about 45 percent of AI news queries produce errors.
Perituza Software Solutions held its second Applying AI in Corporate Workflows workshop, covering sales operations topics including AI agents, CRM integrations, and automated workflows. Attendees from banking, FMCG, and tourism industries also learned to use Microsoft 365 tools like SharePoint and Teams as cost-effective alternatives to third-party SaaS platforms.
AI is gaining acceptance in legal practice for tasks like due diligence, drafting, and contract analysis. The technology helps attorneys analyze large data sets quickly and automate routine document review, though ethical concerns persist around bias and transparency in predictive analytics and risk assessment.
Key Takeaways
- <ul><li>Unsealed court filings reveal OpenAI's datasets contained over 91,692 copies of works from The New York Times and other publications, used for AI training</li><li>Microsoft's Copilot search engine reportedly reduced traffic to The Times' website by up to 93 percent</li><li>Microsoft CEO Satya Nadella testified that paywalled content should be licensed before being used for AI training</li><li>OpenAI called for international AI safety standards and warned against fully autonomous recursive self-improvement without proper safeguards</li><li>The Financial Conduct Authority is considering treating agentic AI the same as human financial advisers, with debate over potential benefits and risks</li><li>IDC predicts 45 percent of G2000 companies will adopt agentic AI supply chain orchestration by 2029</li><li>Nvidia reported $96.22 billion in revenue with 70% expected growth, holding a $5.37 trillion market cap</li><li>About 45 percent of AI news queries produce errors, according to a recent study, prompting Rapid City to warn residents about AI-generated misinformation</li><li>Perituza's corporate AI workshop covered AI agents, CRM integrations, and Microsoft 365 tools as cost-effective SaaS alternatives</li><li>AI is gaining traction in legal practice for due diligence and contract analysis, though bias and transparency concerns remain</li></ul>
New York Times lawsuit exposes OpenAI and Microsoft news scraping for AI training
Newly unredacted court filings in The New York Times' copyright case against OpenAI and Microsoft reveal how news articles were scraped for AI training. The documents show large-scale collection of news content, use of web indexes, and allegations that researchers found ways to bypass The Times' paywall. Microsoft's Copilot search engine reportedly reduced traffic to The Times website by up to 93 percent. The filings also claim OpenAI's datasets contained over 91,692 copies of works from The Times, The Daily News, and the Center for Investigative Reporting.
Unsealed documents show Microsoft and OpenAI executives called AI training theft
Unsealed filings in The New York Times' three-year copyright lawsuit reveal that Microsoft and OpenAI executives privately viewed their AI training practices as theft and an existential threat to publishers. Microsoft's Brent Hecht reportedly called the scraping the largest theft of labor in human history. Microsoft CEO Satya Nadella testified that paywalled content should be licensed before being used for training. OpenAI's ChatGPT head Nick Turley reportedly described the product as increasingly substitutive for publisher content.
OpenAI proposes global AI standards for alignment and recursive self-improvement
OpenAI released proposals for safety and security in frontier AI development, focusing on alignment research and recursive self-improvement, or RSI. The company called for international cooperation to develop frontier standards and recommended building on existing AI safety institutes. OpenAI stated that fully autonomous RSI should not be pursued unless it can be done safely. The proposals were published in a blog post by the ChatGPT maker.
OpenAI outlines standards for the next phase of AI development
OpenAI published a blog post on September 21, 2026 about building standards for the next phase of AI. The post discussed recursive self-improvement, or RSI, where AI systems develop successive generations of AI with increasing automation. OpenAI warned that fully autonomous RSI could result in humans losing practical control over AI development. The company also called for international standards for safety and security practices in frontier AI development to create shared baselines for technical safeguards.
AI in finance needs different regulatory approach, experts say
The Financial Conduct Authority is considering a new regulatory framework for artificial intelligence in finance that would apply the same rules to AI systems as to human financial advisers. Some experts warn that treating agentic AI like human advisers could deprive millions of investors of valuable help. They argue AI systems can provide insights and recommendations that human advisers cannot. The debate continues over how to balance AI benefits with responsible use in finance.
AI agents now manage global supply chain decisions
AI agents are beginning to coordinate supply chain decisions like rerouting ships and reallocating inventory without human approval. These systems handle product, information, finance, and decision flows across global networks. Procurement is seeing the biggest gains from agentic AI because every decision has a measurable financial impact. IDC predicts 45 percent of G2000 companies will adopt agentic AI supply chain orchestration by 2029. Clean governed data remains the key prerequisite for scaling these systems.
Perituza wraps up second AI workshop for corporate sales teams
Perituza Software Solutions finished its second Applying AI in Corporate Workflows workshop focused on sales operations. Professionals from banking, FMCG, tourism and other industries attended the event at Bistro des Marées in Cinnamon Life. The session covered buy versus build decisions, AI agents, CRM integrations, and automated workflows for immediate ROI. Participants also learned how to use Microsoft 365 tools like SharePoint and Teams as cost-effective alternatives to third-party SaaS platforms.
Rapid City warns residents not to trust AI-generated information
Rapid City officials warned residents to verify information after an AI-generated search result falsely claimed the city reimburses property owners for sidewalk and curb repairs. The fake claim blended Rapid City rules with reimbursement policies from Ohio and Indiana. City officials say AI has given wrong information before, including incorrect recycling pickup schedules. A study found about 45 percent of AI news queries produce errors. Officials recommend contacting the city directly for accurate information.
AI gains acceptance in legal practice as strategic tool
AI is now being recognized as a valuable tool in legal practice after earlier skepticism. Attorneys use AI for due diligence, drafting, and contract analysis to free up time for higher-level tasks. The technology can analyze large data sets quickly and automate routine document review. However, ethical concerns remain around bias and transparency in areas like predictive analytics and risk assessment.
AI and data centers dominate Oregon 4th District race
Republican candidate Monique DeSpain and Democrat incumbent Val Hoyle are clashing over AI and data centers in Oregon's 4th Congressional District rematch. DeSpain accused Hoyle of taking money from AI companies, citing $292,000 in independent expenditures from a pro-AI PAC during the primary. Hoyle denied accepting money from AI tech companies and said she supports federal regulation of AI and data centers. DeSpain promised not to accept money from the AI industry but did not detail specific policies.
Nvidia AI growth beats Eli Lilly drug sales for decade potential
Nvidia and Eli Lilly both beat earnings expectations, but Nvidia offers stronger long-term growth. Nvidia reported $96.22 billion in revenue driven by AI infrastructure demand and expects 70% growth next fiscal year. Eli Lilly earned $22.97 billion from GLP-1 drugs like Mounjaro but faces risks from price declines. Nvidia holds a $5.37 trillion market cap compared to Eli Lilly's $1.03 trillion. Nvidia suits growth investors while Lilly fits those seeking dividends and lower volatility.
Sources
- New York Times lawsuit reveals how OpenAI, Microsoft scarped news for AI training
- Unsealed Documents Reveal Microsoft and OpenAI Executives Privately Called AI Training “Theft”
- OpenAI proposes development of global AI standards to guide alignment, RSI
- Building standards for the next phase of AI
- AI in finance must be policed differently
- AI Agents Take the Controls in Supply Chains
- Perituza concludes second AI workshop, helping companies elevate sales execution through AI - Adaderana Biz English
- Rapid City officials warn against relying on AI for accurate information
- AI in legal practice: From skepticism to strategic advantage
- AI, data centers and tech money take center stage in rematch for 4th Congressional District
- Nvidia's AI growth outpaces Eli Lilly's drug sales for next decade potential
Comments
Please log in to post a comment.