The AI boom has produced a new contender. The only question that matters is the price.

Streetwise Economics · Isaac Jonas · May 28, 2026 · 

A note before we begin: I am not a licensed financial advisor. I analyze companies as an independent economist, and everything below reflects my personal analytical views — not a recommendation to buy or sell anything. Anthropic is a private company; its shares are not available on public markets, and the figures here come from public reporting, not audited filings. Markets carry real risk. Do your own research.

A company growing faster than almost any in history

In the last eighteen months, Anthropic has gone from an interesting research lab to one of the most valuable private companies on earth. Its annualized revenue run-rate climbed from roughly $9 billion at the end of 2025 to about $30 billion by April 2026, and independent analysts at Sacra and SemiAnalysis place it near $45 billion by May. For perspective, Salesforce took roughly two decades to reach $30 billion in annual revenue. Anthropic appears to have done it in months.

Most of that revenue is durable in a way investors like: enterprise customers account for about 80% of the total, with more than a thousand businesses each spending over $1 million a year, and over 300,000 business customers in all. Claude Code, the company’s agentic coding tool, reached a $2.5 billion run-rate within months of launch — one of the fastest-scaling software products ever recorded.

So the business is real, and the growth is extraordinary. None of that is in dispute. The investment question is narrower and harder: at what price does that growth become a good deal?

The valuation, and why the multiple is the whole story

Anthropic was last priced at roughly $350 billion in early 2026 (a $380 billion post-money figure on its Series G). There has been credible reporting that the company weighed a later round above $900 billion, and IPO chatter points to a possible listing as early as the fourth quarter of 2026 — though no S-1 has been filed.

Put those numbers against revenue and the picture sharpens. At $350 billion on a ~$45 billion run-rate, you are paying about 8× revenue. At a hypothetical $1 trillion listing, you are paying roughly 22× — for a company that is not yet profitable and has never published audited public financials.

That gap is everything. It is the difference between a demanding-but-defensible price and a price that requires near-perfection just to break even.

What history actually teaches

It is tempting to reach for the great IPO success stories, so let me adjust them honestly rather than cherry-pick. Google went public in 2004 at a $23 billion valuation — about $38 billion in today’s dollars — at roughly 8–10× trailing revenue. Crucially, Google was already profitable. Amazon listed in 1997 at around $438 million (under $1 billion adjusted), at roughly 3× sales, and was unprofitable for years before it justified the faith.

Anthropic at $350 billion sits squarely in Google’s IPO multiple zone — but without Google’s profitability. At $1 trillion, you would be paying roughly twice Google’s IPO multiple for a business still proving its margins can carry the weight. The parallel to history’s winners is real, but it breaks precisely where it matters most.

The risks I price in

Four things keep me cautious. First, Anthropic reports revenue on a gross basis — reselling cloud capacity through AWS, Google and Microsoft inflates the top line relative to net-reporting peers, which means the true revenue multiple is higher than the headline. Second, its compute depends on the very giants — Amazon, Google, Nvidia — that are simultaneously partners, investors, and potential rivals. Third, the entire bull case rests on gross margin climbing from about 50% in 2025 toward a projected 77% by 2028; that operating leverage is forecast, not yet demonstrated. Fourth, there is no audited S-1, no quarterly reporting, and no lock-up history to study.

To be fair to the company: its cash burn is shrinking quickly — to roughly a third of revenue in 2026 and around 9% by 2027 — and management targets profitability by 2028, years ahead of OpenAI. The trajectory is genuinely encouraging.

My scenarios, and my answer

Modeling to 2030 against a hypothetical $1 trillion entry, my bear case lands near $300 billion (a real loss), my base case near $900 billion (barely beating the price), and my bull case near $1.8 trillion. The asymmetry only turns attractive nearer the $350 billion mark — at $1 trillion, you are underwriting the bull case as your base.

So, would I buy if Anthropic listed today? Near $350 billion, I would take a hard, serious look. Near the $1 trillion headline, my honest answer is: wait. I would want the S-1, two quarters of audited public reporting, evidence that margins are expanding on a net basis, and discipline at the lock-up rather than enthusiasm at the bell.

The question was never whether Anthropic is a great company. It may well be. The question is whether the IPO price leaves enough room for the buyer to be right — and still be rewarded.

Isaac Jonas is an economist and the founder of Streetwise Economics. This article is for education and analysis only and is not investment advice.