A private Signal chat full of Silicon Valley's biggest names has spent the last eight months waging a quiet war against a proposed California billionaire tax. Sergey Brin, Marc Andreessen, Chris Larsen, Michael Moritz, and Ron Conway are among the dozens of tech elites who brainstormed everything from buying the signature firm behind the ballot measure to recruiting candidates for governor. So far it hasn't worked. The union backing the tax collected nearly double the signatures it needed, and the billionaires' preferred candidates lost their primaries. Gavin Newsom is now racing to negotiate the measure off the ballot before a June 25 deadline.

China's humanoid robot industry has a new problem, it can build far more robots than anyone is currently buying. Chinese manufacturers now account for roughly 85% of global humanoid robot production, and the government wants 10,000 units in commercial use by the end of 2026. Prices range from about $6,000 for basic models to nearly $100,000 for fully functional ones. Experts say most robots still struggle outside of controlled environments, which means the real test is whether businesses will actually pay for them.

Google just took a Chinese cybercrime ring to court over a massive AI powered scam. The lawsuit accuses a group called Outsider Enterprise of using Gemini to build fake Google, YouTube and government websites to defraud people. The network created 9,000 fake sites and sent millions of scam texts and messages in just two weeks. Google says hundreds of thousands of people were affected, with losses in the millions of dollars. The company is now coordinating with the FBI and pushing Congress to pass new laws against AI driven fraud.

The FBI and the Justice Department shut down thirteen internet domains used by Chinese actors posing as consulting firms. The fake companies used AI generated photos and personas to recruit Americans with security clearances. Job listings for roles like "Senior Analyst" appeared on LinkedIn and other sites, part of a scheme to extract sensitive reports. The action follows a joint FBI and MI5 warning about similar tactics targeting defense analysts.

OpenAI is reportedly weighing steep price cuts across ChatGPT's subscription tiers. The move comes as the company braces for an intensifying fight for users with Anthropic. Both consumer and business plans could be affected, with OpenAI looking to make ChatGPT more competitive on price. It's a sign that pricing is becoming a major new front in the AI assistant wars.

Seattle just became the latest major US city to hit pause on new AI data centers. The city council voted unanimously for a one year moratorium on large facilities, following a wave of public pressure after five new projects were proposed earlier this year. More than 70 cities and counties across the country have now enacted similar bans. New York could be next, with a statewide pause awaiting the governor's signature. It is part of a much bigger story about who pays the cost of the AI boom.

SpaceX is reportedly planning to launch a test of an AI data center in space by the end of 2027. The idea is to put satellites covered in solar panels into orbit, with a central module packed with chips for training and running AI models. It is an ambitious idea, but SpaceX's own IPO filing admits the project might never become commercially viable. Companies have a history of making big promises right before going public, and not all of them pan out.

OpenAI just walked back one of its most ambitious claims. The company had set a goal of fully autonomous AI research by 2028. Now CEO Sam Altman and chief researcher Jakub Pachocki say the future is AI and humans working in tandem, not full automation. They also called for an international body that could slow frontier AI development when societal resilience and safety need time to catch up. The shift arrives as OpenAI is actively preparing for its IPO, which makes the timing worth paying attention to.

Microsoft had a serious security incident this week, and it landed squarely in the AI developer community. Hackers breached dozens of Microsoft's open-source GitHub repositories and injected password-stealing malware into tools used by developers working with Claude Code, Gemini's CLI, and VS Code. More than 70 repos were disabled, and security researchers believe this may be a re-compromise of a project Microsoft thought it had already fixed. It is the second known breach of Microsoft's open-source infrastructure in the span of a few weeks.

UK Prime Minister Keir Starmer walked into London Tech Week and gave Apple and Google three months to block explicit images on children's phones, or face legislation and potential criminal charges for senior managers. The UK's National Crime Agency gets 1,700 child abuse referrals every week, and nine in ten abuse images last year were generated by children themselves, many coerced online. Starmer called it non-negotiable: tech should adapt to society, not the other way around. Civil liberties groups have already pushed back, warning about threats to anonymity and privacy.