OpenAI is a country now

On March 31, 2026, OpenAI announced it had closed a $122 billion funding round at an $852 billion valuation. That's the largest private fundraise in Silicon Valley history, and depending on how you squint at the numbers, it makes OpenAI richer than most sovereign nations. At some point, a company stops being just a company. OpenAI might already be past that point.

Putting $852 billion on the map

To understand how absurd this number is, it helps to stop thinking in market cap terms and start thinking in GDP terms. OpenAI's $852 billion valuation would place it comfortably among the top 20 economies in the world by nominal GDP. It exceeds the entire economic output of countries like Belgium, Ireland, and Switzerland. If OpenAI were a country, it would have a larger "economy" than most of the nations whose governments are now scrambling to regulate it. If it were public today, OpenAI would be the 11th largest company in the S&P 500. Its valuation trajectory tells the story of a vertical line: $28 billion in April 2023, $86 billion by January 2024, $157 billion in October 2024, $300 billion in March 2025, $500 billion by October 2025, and now $852 billion. That's a 30x increase in three years. The comparison to nation-states isn't just rhetorical flair. It's a useful frame for understanding the kind of power that concentrates when this much capital flows into a single private entity.

These aren't investors, they're allies

Look at the roster: SoftBank co-led the round alongside Andreessen Horowitz, D.E. Shaw Ventures, MGX, TPG, and T. Rowe Price. Amazon, Nvidia, and Microsoft all participated. Amazon's remaining $35 billion investment is tied to milestones including a potential IPO by 2028. Nvidia and SoftBank have each committed additional $10 billion installments later this year. This isn't a standard venture capital deal. These are strategic alliances. Amazon gets deeper AI integration. Nvidia secures a customer for its chips at historic scale. SoftBank positions itself at the center of the AI infrastructure buildout. Microsoft, already holding a 27% stake after OpenAI's conversion to a for-profit entity, cements its role as the platform layer beneath the entire operation. The capital isn't passive. It comes with compute commitments, distribution partnerships, and infrastructure agreements. When your investors are also your suppliers, your cloud providers, and your chip manufacturers, the relationship starts to look less like a cap table and more like a treaty.

A cultural phenomenon with a brokerage account

For the first time, OpenAI opened participation to individual investors, raising $3 billion through bank channels. The company is also being included in several ARK Invest ETFs, giving retail investors access to private OpenAI stock ahead of its reportedly upcoming IPO. This is a significant shift. When retail money starts flowing into a private AI company through bank channels and ETF wrappers, it stops being a technology bet and becomes something closer to a cultural phenomenon. People aren't just investing in a product. They're investing in a narrative about the future. OpenAI was the fastest technology platform to reach 10 million users, the fastest to 100 million, and is on track to be the fastest to 1 billion weekly active users. Revenue hit $1 billion annually within a year of launching ChatGPT, reached $1 billion per quarter by end of 2024, and now sits at $2 billion per month. The company claims it is growing revenue four times faster than Alphabet and Meta did in their defining eras. Those are real numbers. But they also exist in the context of a company that is projected to lose $14 billion in 2026.

The governance question nobody can answer

Who controls an entity this powerful? OpenAI started as a nonprofit in 2015, created a capped-profit subsidiary in 2019, and in October 2025 completed its conversion to a Public Benefit Corporation. The nonprofit remains in the structure, technically overseeing the for-profit arm, but the practical dynamics of that oversight are murky at best. The board includes Bret Taylor as chair, alongside Adam D'Angelo, Dr. Sue Desmond-Hellmann, Zico Kolter, retired U.S. Army General Paul M. Nakasone, Adebayo Ogunlesi, Nicole Seligman, Larry Summers, and CEO Sam Altman. A PBC is required to consider interests beyond shareholders, but it is up to the board to decide how to weigh those interests and what to report. The Delaware attorney general approved the restructuring. California signed off. But approval from regulators doesn't answer the harder question: what happens when a PBC with $852 billion in backing, $600 billion in planned compute spending through 2030, and ambitions to build artificial general intelligence decides its interests diverge from the public's? A public benefit corporation is a governance structure. It is not, by itself, a governance solution.

When companies outgrow the word "company"

This isn't the first time a private entity has grown so large that the label stopped fitting. The British East India Company, incorporated in 1600, started as a trade organization and ended up governing large parts of India with its own army, its own courts, and its own tax collection system. At its peak, it was responsible for roughly half of the world's trade. It didn't just participate in markets. It was the market. The Dutch East India Company (VOC), founded in 1602, held a monopoly on the Dutch spice trade and at its peak was worth roughly $7.9 trillion in today's dollars. It functioned as a quasi-sovereign entity, waging wars, establishing colonies, and minting its own currency. Standard Oil, at its height in the early 1900s, controlled roughly 90% of U.S. oil refining. Its influence over the American economy was so total that it took a Supreme Court ruling to break it apart. The pattern is consistent: when a private entity accumulates enough capital, infrastructure, and strategic importance, it begins to function like a state actor, whether or not anyone grants it that status. The question is never whether it will happen. It's whether anyone notices before the lines have already been redrawn.

The structural evolution that tells the real story

OpenAI's organizational history reads like a company trying to outrun its own contradictions. It was founded as a nonprofit with a mission to ensure artificial general intelligence benefits all of humanity. By 2019, it had created a capped-profit subsidiary because the nonprofit structure couldn't attract the capital needed to compete. By 2025, the cap was gone, and the company had converted to a for-profit PBC. The nonprofit received a $130 billion stake in the new entity, a number that sounds like accountability but functions more like a payout. Each structural change was framed as necessary to pursue the mission. And each one moved the company further from the constraints that defined it. The mission statement itself has evolved, with OpenAI reportedly removing the word "safely" from its core mission language. This isn't unique to OpenAI. Anthropic and xAI also use PBC structures. But OpenAI is the one that started as a nonprofit explicitly created to counterbalance the concentration of AI power in corporate hands, and then became exactly the kind of entity it was designed to prevent.

Valuation is not value

It's worth maintaining some skepticism about what $852 billion actually represents. OpenAI generated $13.1 billion in revenue in 2025 and spent $8 billion. It is projected to lose $14 billion in 2026, roughly three times its estimated 2025 losses. The company is targeting $280 billion in annual revenue by 2030, a figure that requires roughly 20x growth in four years. Most of the $122 billion raise is going toward compute. OpenAI plans to spend approximately $600 billion on compute through 2030. Its CFO has publicly stated the company is already turning down opportunities because it doesn't have enough compute capacity. The business model, at its core, is: raise enormous sums of money, convert that money into chips and data centers, and hope that revenue scales faster than costs. That model has worked for exactly one company in history at this scale: Nvidia, which sells the shovels. For the company buying the shovels, the math is less forgiving. None of this means OpenAI will fail. It does mean that the $852 billion number is a bet on a future that hasn't arrived yet, not a reflection of current economic reality.

What this actually means

OpenAI is no longer a startup. It's arguably no longer a company in any conventional sense. It's a capital-absorbing entity backed by the most powerful technology firms and financial institutions on the planet, governed by a structure that was redesigned three times in six years, spending at a pace that rivals national defense budgets, and building technology that its own leadership describes in civilizational terms. That doesn't make it evil. It doesn't make it good. It makes it something we don't have great language for yet. The honest assessment is this: OpenAI has produced genuinely impressive technical work, attracted real users, and generated real revenue. It has also consumed capital at a rate that would be unsustainable for almost any other entity on Earth, restructured its governance to remove the constraints that once defined it, and accumulated a level of strategic importance that makes it effectively too important to fail. When a company's valuation exceeds the GDP of most nations, when its investors are also its suppliers and infrastructure partners, when its governance has been redesigned to accommodate ever-larger capital inflows, the question isn't whether it's a good investment. The question is what kind of institution it has become, and who, if anyone, gets to decide what it does next.