Why Anthropic’s Historic IPO is a Red Flag for Retail Investors
Today, November 19, 2026, Wall Street is celebrating a milestone. Anthropic, the artificial intelligence company behind Claude, is making its debut on the Nasdaq under the ticker ANT at $83.00 per share. Raising $81.3 billion at a staggering $1.5 trillion valuation, it has officially surpassed SpaceX to become the largest Initial Public Offering in history.
To hear the investment banks tell it, this is the dawn of a new economic era. Yet if you look past the champagne and the opening bell, the numbers behind this offering tell a much more sobering story—one that raises serious questions about the financial sustainability of the AI boom and what retail investors are actually buying into.
Paying a $1.5 Trillion Price Tag for Zero Profits
When a company asks the public market to value it at $1.5 trillion—roughly half the market cap of established tech titans like Microsoft, Alphabet, or Nvidia—one might expect a proven, profit-generating machine. Instead, Anthropic enters the public markets without a positive Price-to-Earnings (P/E) ratio because it does not make money.
Over the first nine months of 2026, posted a jaw-dropping $104 billion GAAP net loss. While defenders will quickly point out that roughly 90% of that figure stems from non-cash accounting revaluations on pre-IPO convertible notes, the core business itself is still bleeding cash. Stripping away the accounting quirks, Anthropic recorded a $1.3 billion operating loss over those same nine months, swinging back into the red in Q3 after a brief brush with operating profitability in Q2.
At $83 a share, new investors are paying 22.1 times annualized revenue and over 54 times forward adjusted EBITDA—nearly four times the average multiple of mature software peers. Meanwhile, profitable tech giants like Alphabet and Nvidia trade at a fraction of those earnings multiples while generating tens of billions in actual free cash flow.
The $895 Billion Elephant in the Server Room
The biggest risk to Anthropic’s future—and to the broader AI sector—isn't its past losses, but its locked-in future bills. To train frontier models and keep pace in the AI arms race, Anthropic has signed $895.4 billion in long-term compute and data center lease commitments, including over $417 billion in non-cancellable contracts with cloud giants like Google, Amazon, and Microsoft. That translates to an average infrastructure tab of roughly $100 billion per year looming over the company's balance sheet.
In Q3 alone, surging compute expenses jumped 329% year-over-year, forcing Anthropic to spend $5 billion on short-term compute capacity that wiped out its operating profit. Add in another $23.7 billion in unrecognized stock-based compensation set to hit the books over the next two years, and the path to meaningful bottom-line profitability looks less like a highway and more like a tightrope.
This points to a deeper structural problem across the AI industry: the sheer capital intensity of intelligence. Unlike traditional software—where building a product once allows you to sell it millions of times at near-zero marginal cost—frontier AI requires endless, exponentially expensive computing power just to stay relevant. Even as Anthropic's revenue surges, its costs are scaling right alongside it.
Who Really Wins When the Bell Rings?
In this historic debut, up to 20% of the IPO shares were carved out specifically for individual retail investors. Wall Street is framing this as democratizing access to the future of technology. In reality, it looks far more like an exit ramp for early insiders.
Consider the timeline of Anthropic's private market valuations:
February 2026: Private investors bought in at a $380 billion valuation.
May 2026: Secondary shares pushed the valuation to $965 billion.
November 2026: The public is invited to buy in at $1.5 trillion.
Early institutional backers from just nine months ago are already sitting on nearly 4x paper gains at the $83 offer price. Even the major tech partners buying into the $26 billion concurrent private placement—Google, Nvidia, and Microsoft—aren't purely betting on stock appreciation; they are the very cloud and chip vendors set to collect hundreds of billions of dollars from Anthropic’s locked-in compute contracts.
The Bottom Line for New Investors
Buying at $83 isn't investing in a proven business model; it is making a high-stakes wager that the company can achieve a rate of sustained, hyper-exponential revenue growth that almost no company of this size has ever pulled off—all while paying off an unprecedented $895 billion infrastructure tab.
History shows that transformative technologies often create massive value for society long before they create reliable profits for late-stage shareholders. As the tickers flash green on the Nasdaq today, retail investors would do well to ask themselves whether they are buying a stake in the future of AI, or simply footing the bill for its most expensive experiment yet.
.png)













Comments