AI Reporter – August 2026

AI Update

Intellectual property disputes continue to predominate in AI litigation as courts and litigants increasingly focus on model training practices, evidentiary obligations and the scope of fair use. Midjourney asked a California federal court to compel Disney, Universal and Warner Bros. to produce broader discovery regarding their own internal use of generative AI, arguing those materials could undermine the studios’ claims of competitive harm and copyright infringement. Copyright litigation also intensified against AI music companies, with Sony filing a new lawsuit accusing Udio of infringing more than 30,000 recordings, and reports that hacked Suno source code identified YouTube Music, Deezer and other online services as sources of training material. News organizations led by The New York Times sought sanctions against OpenAI, alleging the company failed to preserve evidence and misrepresented its ability to identify copyrighted material within its training datasets and ChatGPT outputs.

AI-related litigation also continues to expand beyond intellectual property into employment, privacy and platform liability, reflecting increasing judicial scrutiny of how AI systems are developed and deployed. Apple sued a former engineer who joined OpenAI, alleging he exploited a software flaw to retain access to Apple’s internal systems and downloaded confidential hardware designs and manufacturing information before departing.

On the regulatory front, governments are increasingly shifting from broad AI policy discussions toward targeted oversight addressing infrastructure, employment, consumer protection and national security. Illinois enacted new AI accountability legislation while separately prohibiting the use of AI-generated scores in teacher evaluations. Connecticut imposed new disclosure and compliance obligations governing employers’ use of AI-assisted hiring tools, and North Carolina released a statewide AI road map emphasizing governance, consumer safeguards and workforce development.

Federal policymakers are also pursuing more comprehensive oversight, with Senator Mark Warner introducing legislation that would require security testing for advanced AI models, establish consumer protections for AI agents, and expand workforce transition programs, while Congress separately introduced the AI Kill Switch Act requiring developers to maintain the technical ability to suspend or disable powerful models following recent cybersecurity concerns.

International competition continues to shape AI governance as U.S. and Chinese officials prepare bilateral talks on frontier-model risks, France aggressively courts AI infrastructure investment, and China warns it will retaliate against potential U.S. sanctions targeting Chinese AI developers.

These and other stories appear below.

AI in Business

Orchestra expands AI-powered street camera network across San Francisco

A new start-up has deployed more than 100 street-facing and continuously filming cameras across San Francisco and plans to add 900 more over the next 6 months. Start- up Orchestra uses AI to convert the footage into structured data on objects, vehicles and incidents. The company says it will sell the data to a variety of customers, including police departments, insurance providers, autonomous vehicle companies and real estate firms, highlighting growing commercial use of AI-enabled surveillance and analytics in urban settings.

Source: Business Insider (sub. req.)

Microsoft commits $2.5 billion to new enterprise AI deployment unit

Microsoft launched a new operating business focused on enhancing the company’s use of enterprise AI through existing Microsoft tools. The Microsoft Frontier Company is backed by a $2.5 billion investment and 6,000 industry and engineering experts. Amazon Web Services, OpenAI, and Anthropic have announced similar AI deployment ventures. Microsoft said early partnerships include London Stock Exchange Group, Unilever, Land O’Lakes and Accenture as it leverages its existing relationships across the Fortune 500.

Source: TechCrunch

OpenAI proposes 5% U.S. government stake amid Trump AI pressure

OpenAI has reportedly discussed giving the U.S. government a 5% ownership stake. Chief Executive Officer Sam Altman is said to have pitched the idea to President Donald Trump as a way of giving the public a financial interest in AI’s success while reducing political and public pressure on the sector. Based on OpenAI’s latest funding round, which values the company at $852 billion, the stake would be worth about $42.6 billion.

Source: The Verge (sub. req.)

Reddit deploys LLM-based spam detection as AI-generated bot content rises

Reddit is using AI to combat the growing volume of spam and bot content enabled by widely accessible large language models (LLMs). The company said its updated spam-fighting system blocks 23 million spam views per day and catches about 25,000 new spam posts and comments at the same time. According to Reddit, the LLM-based tools reduced users’ exposure to spam by 20% from January to March, compared with the prior three months.

Source: TechCrunch

Alibaba bans Anthropic AI tools for employees after distillation attack accusation

Alibaba will prohibit employees from using Anthropic AI tools starting July 10. The company previously classified Claude Code as high-risk software over alleged back-door security risks. Employees are reportedly being told to uninstall Anthropic models and agent products and to use Alibaba’s own AI assistant instead. The move follows an AI-related dispute between the two companies in June. Anthropic had written to the U.S. Senate Committee on Banking, Housing and Urban Affairs, accusing Alibaba of “brazenly” and “illicitly” attempting to extract its AI capabilities through what Anthropic called the “largest known distillation attack” against the company to date.

Source: CNBC

Hollywood animators face rising AI disruption risk as jobs contract

Animators, visual effects artists, and concept and storyboard artists are among the Hollywood roles seen as most vulnerable to AI-related disruption, according to an industry survey. The survey revealed workers are already receiving fewer opportunities amid a broader downturn driven by post- pandemic production shifts, fewer green-lit projects and labor strikes.

Source: The Atlantic (sub. req.)

xAI installed 59 unpermitted gas turbines for Tennessee AI data center

Elon Musk’s AI company installed 59 natural gas turbines for its Colossus 2 data center project in Tennessee without obtaining federal clean air permits, according to communications between regulators and the company’s representatives. xAI has previously said permits were not required and had publicly acknowledged operating 27 unpermitted turbines as of January, although potential emissions appear to exceed the threshold requiring federal permits.

Source: Reuters (sub. req.)

Meta and BlackRock launch $14B El Paso data center venture for AI computer

Meta and BlackRock announced plans to develop and operate a data center campus in El Paso, Texas, as demand for AI infrastructure drives large-scale capital raising. BlackRock-managed funds will hold an 80% stake in the project and finance part of the investment with $12.5 billion in debt, while Meta will retain 20%, receive a $1 billion distribution and contribute land and construction assets worth about $2.3 billion. The campus is designed to provide 1 gigawatt of compute capacity for Meta’s AI technologies and will cost about $14 billion in development.

Source: Reuters (sub. req.)

To view the full article click here