The law and the money caught up to AI in the same month

This month AI stopped being a story about models. It became a story about money and law, and the two turned out to be the same story. Everything below points at one rule. Control what you depend on.

Business

The money around AI got hard to picture.
Anthropic raised 65 billion dollars at a 965 billion dollar valuation, with run rate revenue past 47 billion, and passed OpenAI as the most valuable AI startup. OpenAI raised 122 billion at an 852 billion valuation. Both filed quietly to go public. Venture funding hit 300 billion dollars in a single quarter, and about 88 percent of it chased AI. In the same stretch, 2026 tech layoffs passed 156,000 people, including 21,000 at Oracle, much of it blamed on AI.

The play: the capital and the cuts are telling you the same thing. Stay lean and let the system do the work, so you are the one cutting cost, not the cost being cut.

Technology

Three frontier models shipped in about five weeks.
OpenAI's GPT-5.5 landed April 23. Google's Gemini 3.5 Flash on May 19. Anthropic's Claude Opus 4.8 on May 28, at the same price as the model before it, with a new dynamic workflows mode for large jobs. NVIDIA put its next chip platform, Rubin, into full production to run all of it.

The play: capability is no longer the bottleneck. Pick one model stack, wire it into real work, and ship. Your edge is what you build on top, not which model you picked this week.

Legal

Anthropic, now worth 965 billion dollars, is paying 1.5 billion for how it trained.
Anthropic settled with authors for 1.5 billion dollars, the largest copyright settlement in US history, near 3,000 dollars a book. A judge had ruled that training on books was fair use, but downloading millions of pirated ones was not. Music publishers led by Universal Music Group are now suing Anthropic for more than 3 billion over song lyrics. Reddit is suing Anthropic over scraped user data and just won the right to fight in state court. The New York Times forced OpenAI to hand over a 20 million chat log sample. Two separate Google antitrust cases could reshape search and ad tech this year.

The play: how you source your data is now a balance sheet item, not a footnote. Keep proof of where your data came from before anyone asks.

Politics

On August 2, the EU can start fining AI model makers up to 3 percent of global revenue.
That is the day the EU AI Act's enforcement teeth turn on for general purpose models, with fines up to 3 percent of worldwide turnover or 15 million euros. In the US it runs the other way. A December executive order is trying to override state AI laws, while California, Texas, and Colorado put their own rules into force during 2026. Where you operate now decides which rulebook applies to you.

The play: choose your jurisdiction on purpose. The map of where a business is treated best now includes its AI rules.

Other signals

Anthropic passed OpenAI as the most valuable AI startup, 965 billion to 852 billion.
OpenAI is making about 2 billion dollars a month and still does not expect to be cash positive until 2030.
Databricks and SpaceX are both lined up for large IPOs behind Anthropic and OpenAI.
Sony is the last major label still suing Suno and Udio, after Universal and Warner settled.
Colorado's AI Act takes effect June 30, the first US law aimed at algorithmic discrimination in high risk systems.

One thing to keep

Control what you depend on. The model you build on, the data you train on, and the rules you operate under are each a lever someone else can pull. Own your provenance, your distribution, and your jurisdiction before they own you.

Built from nothing, operated from anywhere.
Chase and Kobe

This is why I am building NomadLaw.ai, to help people stand up to predatory institutions that use intimidation and aggressive litigation to wear them down, anywhere on earth. Not legal advice.

If this saved you an hour of scrolling, forward it to one founder who is building.

Not legal, financial, or tax advice.