What is Anthropic vs. OpenAI: The First-Ever AI IPO Race Is On — Two Frontier Labs File to Go Public Within a Week?
Within the span of one week, the two companies that have come to define the artificial-intelligence boom both did something they had spent years insisting they were in no rush to do: they filed to go public. Anthropic submitted confidential IPO paperwork on June 1, 2026, at a staggering $965 billion valuation. Seven days later, on June 8, OpenAI quietly followed with its own confidential filing. For the first time since the AI race began, the finish line is not a better model or a bigger funding round — it is the opening bell of a stock exchange, and the two labs are sprinting toward it side by side.
The timing has turned what was already a fierce technical rivalry into a full-blown Wall Street spectacle. Anthropic has, for the first time, overtaken OpenAI on paper, and it did so days before shipping a new flagship model that promptly topped nearly every benchmark in the field. This is the week the AI arms race stopped being purely about technology and became a question every investor, employee, and customer now has to answer: which of these two companies is actually worth more — and is either of them worth what the private markets say?
Why the AI IPO race is suddenly everywhere
The story is trending right now because three separate threads collided in the first two weeks of June 2026, and each one would have been a headline on its own.
First, the filings themselves are historically unusual. Confidential IPO paperwork lets a company begin the listing process without immediately disclosing financials to the public, and both Anthropic (June 1) and OpenAI (June 8) used that mechanism within a week of each other. The result is a genuine race: two frontier AI labs targeting public debuts in the same fall-2026 window, something that has simply never happened in this industry. OpenAI is reportedly working with Goldman Sachs and Morgan Stanley toward a September debut at a valuation in the $730 billion to $850 billion range.
Second, the valuation crown changed hands. Anthropic's $965 billion figure — set in a Series H round that raised roughly $65 billion — pushed it past OpenAI's $852 billion post-money valuation from March, making Anthropic the most valuable startup in the world for the first time. On the secondary market Forge Global, Anthropic shares even briefly touched a $1 trillion mark. The symbolism is enormous: the younger, smaller company, founded by ex-OpenAI researchers, is now ahead on the one scoreboard Silicon Valley watches most closely.
Third, the product timing could not have been sharper. Just as the financial headlines landed, Anthropic released a new top-tier Claude model on June 9 — Claude Fable 5 — that it called state-of-the-art on nearly every benchmark it tested. Shipping a record-setting model in the same week you file to become the most valuable company in your sector is the kind of one-two punch that sends a story vertical, and search interest in both companies' valuations spiked accordingly.
It is worth pausing on just how strong that Fable 5 launch was, because it underpins the valuation argument. Anthropic reported the model posting 80.3% on SWE-Bench Pro and 88.0% on Terminal-Bench 2.1 — coding-and-agent benchmarks that translate directly into the enterprise developer use cases driving its revenue — and a category-leading score on its own real-world agentic-work benchmark. Anthropic also cut the price relative to its previous top model, a signal it intends to compete on cost as well as capability. For investors trying to underwrite a near-trillion-dollar valuation, a frontier model that is both better and cheaper is exactly the proof point a fall IPO roadshow will lean on.
How Anthropic pulled ahead
For most of the AI era, the running order felt settled: OpenAI was the household name, the company with ChatGPT and 900 million weekly users, and everyone else was chasing. So how did Anthropic, founded only in 2021, vault past it on valuation?
The short answer is revenue velocity, and a particular kind of revenue at that. Anthropic's annualized run rate has gone from roughly $87 million in early 2024 to figures reported around $30–40 billion by spring 2026 — a growth curve so steep the company itself described it as "crazy." Crucially, much of that growth is enterprise and developer spending, the kind investors prize because it tends to be sticky and high-margin. The engine has been Claude Code, Anthropic's agentic coding tool, which reportedly hit $1 billion in annualized revenue within six months of launch and crossed $2.5 billion in run-rate by early 2026, making it the fastest-growing product in the company's history.
The enterprise footprint underneath those numbers is substantial: Anthropic has said it serves more than 300,000 business customers, has grown its number of large accounts roughly sevenfold in a year, and now counts more than 500 customers spending over $1 million annually. That mix matters to public-market investors, who tend to reward durable business spending over consumer subscriptions that can churn. We traced the start of this momentum back in our coverage of Anthropic's earlier $30 billion round at a $900 billion valuation, and the June filing is the next chapter of that same arc.
There is also a profitability story. By Anthropic's own projections, it expects to break even around 2028 — roughly two years ahead of OpenAI's stated 2030 target. In a market suddenly being asked to price these companies as public stocks rather than venture bets, a shorter path to positive cash flow is a meaningful differentiator.
There is a structural difference too. OpenAI spent much of 2024 and 2025 building a sprawling consumer business — a free tier with hundreds of millions of users, a subscription layer, and an advertising-adjacent surface area that is expensive to serve and harder to monetize per user. Anthropic, by contrast, deliberately leaned into developers and enterprises from the start, which means a larger share of its revenue comes from customers paying by usage rather than a flat monthly fee. Public-market analysts tend to assign richer multiples to that kind of consumption-based, expansion-friendly revenue, and it helps explain why a company with fewer total users can command a higher valuation than the one with the more famous chatbot.
What OpenAI brings to the fight
None of this means OpenAI is a fallen giant. By raw reach, it remains the most consumer-dominant AI company on the planet. ChatGPT crossed roughly 900 million weekly active users by early 2026, and OpenAI's revenue — around $2 billion a month, or roughly $25 billion annualized as of early 2026 — dwarfs almost everyone outside Anthropic. Its March 2026 round committed an enormous $122 billion in capital at that $852 billion valuation.
OpenAI's pitch to public investors is scale and ubiquity: a brand that has become synonymous with AI itself, an estimated 92% of Fortune 500 companies using its products in some form, and deep distribution through partners like Microsoft, which holds a roughly 27% stake in OpenAI's for-profit unit. For retail investors who cannot currently buy either company directly, that Microsoft stake has been one of the only ways to get indirect exposure — a workaround that an actual IPO would finally render unnecessary.
But OpenAI also carries the heavier financial narrative. By its own projections, the company is not expected to generate more cash than it spends for several more years, with profitability targeted around 2030. It is spending ferociously on compute and infrastructure to keep its models at the frontier, the same dynamic now reshaping the broader market for AI agents in knowledge work. When it goes public, OpenAI will be asking Wall Street to fund years of planned losses on the bet that its lead in consumer mindshare eventually converts into commensurate profit.
The first real test of AI valuations
The deeper reason this race matters goes beyond either company. Until now, AI valuations have been set almost entirely in private — by venture firms, sovereign funds, and secondary markets, where prices can run on narrative and FOMO as much as fundamentals. An IPO changes the rules. Public companies must file detailed financials, and public investors price them every second the market is open. For the first time, the AI boom is about to be marked to market.
That is why analysts keep calling these filings the first big stress test of AI-era valuations. If both Anthropic and OpenAI debut in the fall, investors will be able to compare two frontier labs side by side using real, audited numbers — revenue, losses, margins, growth rates — rather than pitch decks. The outcome could validate the entire framework that has lifted AI valuations into the hundreds of billions, or it could expose just how much of those numbers rests on projections that have not yet materialized. The stakes ripple straight into the index funds and ETFs that millions of ordinary investors hold, since a handful of AI-exposed names already drive an outsized share of major market returns.
The concentration of these offerings is itself remarkable. Anthropic, OpenAI, and SpaceX — the last of which began trading on June 12 — together represent a pipeline that could demand well north of $100 billion from public markets in a matter of months. For context, the entire U.S. IPO market raised roughly $45 billion in all of 2025. We broke down the rocket-company side of that wave in our look at the record-setting SpaceX IPO; the AI labs are the other half of the most concentrated cluster of mega-listings the market has seen since the dot-com era.
What it means for everyone else
For developers and businesses already building on these platforms, the IPO race introduces a new variable. Public companies answer to quarterly earnings and shareholder pressure in ways private labs do not, which can change how aggressively they price products, how they handle safety trade-offs, and how patient they can afford to be. The model wars will not slow down — if anything, the need to show growth to public investors may accelerate the release cadence that already has new frontier models arriving almost monthly, a pace we tracked when four AI labs made four acquisitions in five days.
For retail investors, the practical takeaway is patience. A confidential filing is the beginning of a process, not the end; actual shares may not trade until the fall, and even then, the first frontier-AI IPOs are likely to be volatile as the market figures out how to value businesses growing this fast while spending this heavily. Buying into the hype at any price has rarely been a winning strategy, and anyone weighing exposure would do well to understand the difference between owning a piece of a profitable, enterprise-anchored business and bankrolling a consumer juggernaut's years of planned losses. For those tempted to chase adjacent speculative bets, the same caution applies across crypto and other high-volatility corners of the market that tend to move in sympathy with AI sentiment.
What is no longer in doubt is that the era of AI-as-a-private-experiment is ending. Two of the most consequential companies of the decade are about to open their books to the world, and a third — SpaceX — already has. By the time the leaves turn this fall, the question "how much is AI really worth?" will finally have a number the whole market can see, argue over, and trade on. For ongoing coverage of how the AI race and the broader AI tools landscape keep reshaping technology and markets, follow our continuing reporting at trends.thicket.sh.
Origin
Anthropic confidentially filed for an IPO on June 1, 2026 at a $965 billion valuation (after a ~$65 billion Series H), surpassing OpenAI's $852 billion March 2026 valuation and becoming the world's most valuable startup. OpenAI filed its own confidential IPO paperwork on June 8, 2026 (with Goldman Sachs and Morgan Stanley), targeting a fall debut around $730-850 billion. Anthropic released Claude Fable 5, a state-of-the-art Mythos-class model, on June 9. SpaceX began trading June 12. Verified via Fortune, CNBC, TechCrunch, Morningstar, and Anthropic's newsroom.
Timeline
Why Is This Trending Now?
Three threads collided in the first two weeks of June 2026. (1) Historic filings: Anthropic filed confidentially for an IPO on June 1 and OpenAI followed on June 8 — the first time two frontier AI labs are racing toward public debuts in the same fall-2026 window. (2) The valuation crown changed hands: Anthropic's $965 billion valuation (and a brief $1T mark on Forge Global) overtook OpenAI's $852 billion for the first time, making Anthropic the world's most valuable startup. (3) Product timing: Anthropic shipped its benchmark-topping Claude Fable 5 model on June 9, the same week as the financial headlines, sending search interest in both companies' valuations vertical — alongside SpaceX beginning to trade on June 12.
Frequently Asked Questions
Sources
- Fortune – Anthropic confidentially files for IPO at $965 billion valuation
- Morningstar – Anthropic Bests OpenAI in Valuation Race, Hitting $965B
- TechCrunch – Following Anthropic, OpenAI files confidentially for IPO
- CNBC – OpenAI confidentially files for IPO, prepping Wall Street for AI debut
- CNBC – Anthropic tops OpenAI as most valuable AI startup
- CNBC – The Tech Download: Anthropic's IPO sets up first big test of AI boom valuations
- Anthropic – Claude Fable 5 and Claude Mythos 5
- VentureBeat – Anthropic hits $30 billion revenue run rate after 'crazy' 80x growth
- AI Magazine – What OpenAI and Anthropic IPOs mean for the AI industry
