PBC’s landmark initial public offering is laying bare the stark extremes of the AI boom business model: astonishing growth but deep uncertainty about unpredictable demand and the massive spending required to continue building the resource-intensive technology. The startup, which publicly filed IPO paperwork on Monday, is poised to be the first major US firm focused solely on building AI software to make its market debut.

The filing offers proof of the billions of dollars in sales per year that Anthropic's products generate, but also puts forth hard questions about its extreme losses, profitability, government intervention and the potential for catastrophic harms that are specific to cutting-edge AI developers.

Just five years into its existence, Anthropic is expected to raise more money in its IPO than any firm in history and overtake Berkshire Hathaway Inc. in market value. Its valuation has surged from $183 billion a little over a year ago to $965 billion in its most recent round, with early expectations that the company could be worth upwards of nearly $2 trillion once it goes public — creating a massive windfall, even for employees and investors who got into the company somewhat late.

Its public market performance might cement or capsize Wall Street’s confidence in the trillion-dollar AI boom at a moment when investors are increasingly anxious about the borrowing binge underpinning it. More than any other companies, Anthropic and archrival OpenAI have set the pace for global AI development. Anthropic has also put itself at the center of a complex web of interconnected investments from leading chipmakers and cloud providers that are helping to prop up the entire sector and the US economy.

“This will really be investors’ first significant opportunity to invest directly in the company that is driving the AI economy,” said Gil Luria, head of technology research at DA Davidson. “It’s of monumental importance.”At the same time, Anthropic’s IPO may shape the prospects for an entire generation of AI startups. It’s anticipated to funnel more capital into the ecosystem through what promises to be an almost unparalleled wealth creation event and is expected to open the floodgates for others to go public in the coming months after a period of paralysis for the IPO market.

In a sign of Anthropic’s influence, investors made the pilgrimage to its San Francisco offices this month to hear executives lay out their vision while a small group of protestors gathered outside, calling to “block the IPO” over concerns about the technology. Public and private markets have closely scrutinized every tidbit of information on Anthropic’s business, including its non-standardized revenue figures.

And on the eve of Anthropic staffers becoming millionaires, wealth management firm IEQ Capital threw a party for AI investors and employees, according to one person familiar with the matter. The invite-only party promised a performance from DJ Diplo.

“Never have IPOs represented so much to the venture capital industry writ large,” said Ravi Viswanathan, founder and managing partner of NewView Capital, about the much-anticipated offerings of Anthropic and, eventually, OpenAI. Over the years, Wall Street has grown accustomed to a certain Silicon Valley success story. Tech firms, including Amazon.com Inc. and Meta Platforms Inc., spend money to increase their share of the market, then reach an inflection point where they become consistently profitable.

Anthropic, by its own account, isn’t promising any consistency. “I don’t think the economics of this industry work that way,” Chief Executive Officer Dario Amodei said in a lengthy podcast interview near the start of the year. It’s uniquely challenging to predict future demand for a rapidly evolving technology and pace costly, longer-term computing investments accordingly, he said. “If I’m just off by a year in that rate of growth, or if the growth rate is 5x a year instead of 10x a year, then you go bankrupt.”

Founded in 2021 by former OpenAI employees with the goal of building AI more responsibly, Anthropic now embodies the core tensions that have come to define the industry. It’s simultaneously calling for slowing down the pace of AI development to mitigate the potential for severe harms to humanity and pushing ahead with a public offering that’s likely to create added shareholder pressure to generate greater returns on investment. Some, including New York City’s comptroller, previously called on Anthropic to pause its IPO until it fully addresses safety risks.

Anthropic intends to use the capital raised from its public offering to fuel investments in chips, data centers and talent to build a technology it believes can reshape the global economy.

Though that goal is lofty and far from certain, Anthropic has gained momentum this year thanks to a surge in demand for its Claude AI software to streamline more complex workplace tasks, particularly writing and debugging code. Like other AI firms, however, it is working to prove the broader utility for its products beyond software engineers, who are typically early adopters.

Anthropic, which had over 1,000 percent revenue growth year-over-year in 2025, has since seen that pace temper. The company’s revenue hit $16.3 billion in the first half of 2026, according to its IPO filing, a more than 14-fold increase from same period in the prior year. But as Bloomberg previously reported, its revenue rate of growth has since slowed from around 140 percent to 45 percent in the second to third quarter in 2026.

The company also faces a concentration of risk: In the first half of 2026, 48% of Anthropic’s sales to customers came from three technology firms’ cloud marketplaces, some of which are building their own models and may someday wean off of Claude, threatening future sales.

Meanwhile, Anthropic disclosed wild swings in its bottom line, including a net loss of almost $42 billion in 2025 and nearly $81 billion in the first two quarters of this year, easily dwarfing the amount Uber Technologies Inc. burned through annually before going public. And the Claude chatbot maker said that just since the end of 2025, it has committed to invest a staggering $417 billion, with terms ranging from one year to 20, on computing and infrastructure capacity to support its software.

A core assumption is that Anthropic and its peers will eventually generate stratospheric revenue from more advanced AI software in the years ahead that offsets the immense investments made today. But no one, including Amodei, knows for sure what sales will be in five years, said Aswath Damodaran, professor of finance at New York University’s Stern School of Business. “They are just as lost as you or me.”

Amodei and Anthropic have repeatedly been blindsided by events this year: A massive surge in demand for its products forced leadership to scramble to line up enough computing resources to keep services running. An unprecedented order from the Trump administration led the company to briefly disable access to its best technologies. And a groundswell of concerns from its own rank-and-file staffers pushed Amodei and Anthropic to publicly reckon with the existential risks of the technology, prompting new confrontations with the White House.

Over the course of almost 100 pages, Anthropic laid out a long list of unique risk factors, ranging from heightened government scrutiny that could slow or derail its ability to release more powerful models — the engine for its entire business — to the possibility that its technology leads to the extinction of the human race.

It also faces a crowded field of competitors that are constantly one-upping each other with more powerful and affordable models, including Chinese firms that have allegedly extracted results from Anthropic’s software to gain an edge and undercut US rivals on price. The combination of novel, existential risks and the fact that Anthropic remains a relatively young company make it harder for investors to model the upsides and downsides of the business, according to advisers close to the IPO, who spoke on condition of anonymity for fear of reprisals.

Still, as a public company, Anthropic’s stock is likely to become the ultimate barometer for the AI boom. Even before its IPO filing, the company’s every move was scrutinized by investors. An understated announcement about legal plug-ins helped spark a meltdown for software stocks. Its revenue is closely watched to gauge demand for AI services broadly. Amodei’s prognostications have been widely dissected, too. A former academic, Amodei has released several lengthy treatises laying out his vision for the benefits and risks as AI advances to transform society with what he called “a country of geniuses in a data center.”

In multiple essays, however, he has acknowledged the challenge of predicting what’s to come. “The effects of powerful AI are likely to be even more unpredictable than past technological changes, so all of this is unavoidably going to consist of guesses,” he wrote in Machines of Loving Grace, a widely-read essay later reprinted in pocket-size bound copies given out to employees. “But I am aiming for at least educated and useful guesses.”