Money is a story we agree to believe. A dollar buys a dollar’s worth because we all act as if it does, and a company is worth whatever the next buyer will pay, not a penny more.
For most of the past century, the biggest stores of that belief were countries and the giant public companies their citizens could actually own a slice of. You could buy Coca-Cola, your local bank, or the carmaker down the highway, and the value sat in plain sight on a stock exchange.
That arrangement is quietly breaking. Some of the most valuable enterprises on Earth are now private, held by venture funds and insiders rather than the public, and ordinary investors stay locked out until the company decides to let them in.
The hottest of them all just took two steps that bring it closer to your brokerage account.
Anthropic, the artificial intelligence (AI) lab behind the Claude chatbot, confidentially filed for an initial public offering (IPO) on June 1.
A day later, it scaled its most powerful model to critical infrastructure across more than 15 countries.
How AI startup Anthropic got bigger than most national economies
Anthropic, founded in 2021 by a group of researchers who left OpenAI, raised $65 billion in a Series H round on May 28.
The deal valued it at about $965 billion and pushed it past OpenAI as the most valuable AI startup in the world, according to CNBC.
More AI:
- Micron sits at the center of a red-hot chip rally
- IBM CEO sends blunt message on AI and quantum computing
- Anthropic CEO makes shocking admission about AI
The filing drops Anthropic into a three-way sprint for the public markets. SpaceX is expected to list first, with Anthropic and OpenAI racing to be the second company ever to go public near or above a $1 trillion valuation, Axios reported.
When I ran Anthropic’s valuation against the IMF’s 2025 GDP table, the AI lab landed around the world’s 20th-largest economy, bigger than the yearly output of entire nations.
- Anthropic’s roughly $965 billion valuation tops the 2025 gross domestic product (GDP) of Belgium, Sweden, or Argentina, based on IMF figures.
- Only about 21 countries produced more than $1 trillion in goods and services in 2025, according to the IMF.
- Anthropic’s revenue run rate reached about $47 billion in May, up from roughly $10 billion a year earlier, CNBC noted.
A valuation is not the same as GDP. One measures what investors will pay for a piece of a company, while the other measures a full year of a country’s output. Even so, the comparison captures something real about where wealth is pooling, and how few hands hold it.
Why Anthropic scaled its most powerful model the day after filing
The model at the center of the second announcement is Claude Mythos, which Anthropic has called its most powerful system, able to find thousands of zero-day software flaws over a matter of weeks, according to TechCrunch.
On June 2, the company widened Project Glasswing, its effort to find and fix critical software vulnerabilities, to about 150 organizations across more than 15 countries, TechCrunch confirmed. An earlier cohort of 50 partners, including the U.S. government, got access in April.
The newly added partners run power grids, water systems, hospitals, communications networks, and hardware supply chains. A major attack on some of them could affect “more than 100 million people,” Anthropic said.
Related: Anthropic drops new Claude model as OpenAI IPO race heats up
The expanded group reportedly includes identity-security firm Okta (OKTA), South Korea’s Samsung and SK Hynix, the NATO military alliance, and the European Union’s cybersecurity agency, the Financial Times reported.
In my reading, the timing is the real story. Rolling out a flagship security model the morning after an IPO filing is a confidence signal aimed straight at the investors Anthropic now needs to win over.
It also lands in a fight. Rival OpenAI has released its own cybersecurity model, GPT-5.5-Cyber, to a group of testers, TechCrunch indicated. The two firms are “in a race to go public before capital runs out,” DA Davidson analyst Gil Luria said, according to Al Jazeera.
What a private trillion-dollar bet means for your wallet
Here is the part that touches your money directly. You cannot buy Anthropic shares today, and no pricing date has been set.
But you already own its world. Claude runs on chips from Nvidia (NVDA) and cloud capacity from Amazon (AMZN), Alphabet (GOOGL), and Microsoft (MSFT), the kind of stocks parked in most index funds and 401(k) plans. When Anthropic finally prices its IPO, it could reset how the entire AI complex gets valued, your holdings included.
There is a quieter payoff, too. The same kind of model your kids will grow up alongside is now scanning the code that keeps hospitals and power grids running.
Whatever you make of the trillion-dollar math, the machines have stopped just chatting. They have started guarding the lights, and the next test is whether public investors will pay a country’s worth to own them.
Related: Anthropic just landed one of the biggest deals in AI
