NYU Stern finance professor Aswath "The Dean of Valuation" Damodaran has a message for anyone expecting Anthropic's IPO to be a careful exercise in intrinsic value: don't hold your breath. In his view, the offering will be priced, not valued. Market dynamics, not some spreadsheet-derived notion of business worth, will set the number.

At an October 15 pre-IPO meeting in San Francisco, Anthropic reportedly posted third-quarter revenue of $16.8 billion and an annualized revenue run rate (ARR) of $67 billion, with sights set on $100 billion by year-end.

Damodaran sees a familiar pattern here. In the late 1990s, dot-com companies were priced on website visitors and then on "eyeballs," a measure of time spent on sites. Now he sees ARR as the AI era's version of that metric: a creative yardstick for companies whose financial statements can't support their market prices.

Young Growth at Massive Scale

Under Damodaran's life-cycle framework, Anthropic is a young growth company that happens to be enormous. Its very high revenue growth, operating losses, heavy reinvestment and negative cash flows are "features not bugs."

However, he dismisses the company's claim of an adjusted third-quarter profit.

"I don't think even Anthropic believes it made money," he said, calling adjusted figures "often a fictional number." Free cash flow remains deeply negative.

Capital intensity is a big factor. Anthropic has more than $400 billion in contractual compute commitments. Since the firm is still losing money and operates in a business environment still just under formation, those obligations are real risks while investors have to price potential rather than proven cash flow.

The Murky Mathematics of ARR

Damodaran writes that ARR is "not standardized and can be easily gamed." Companies can annualize from a day, a week, or a month of revenue. Both AI leaders currently use monthly figures, but "there is nothing that binds them to continue to do so."

Accounting choices make comparisons harder. OpenAI switched to net revenue reporting, which cut its ARR from $70 billion to $50 billion. Anthropic still reports gross revenue and books cloud-provider payments as expenses.

At $1.5 trillion, Anthropic trades at about 23 times current ARR, but only five times a projected $300 billion in 2030 revenue.

"You're going to see a lot of Anthropic bulls use forward numbers. Really, really forward numbers," he predicted, "because it makes your metrics look more reasonable."

The Peer Group Dilemma

Pricing depends on comparable companies, and almost no pure-play public LLM companies exist to compare against.

One approach pairs Anthropic with OpenAI, another private firm, so comparisons rely on stale venture rounds. The broader approach compares the firm with the Magnificent Seven and Space Exploration Technologies Corp. (NASDAQ:SPCX), although these sharply differ in maturity, governance, and cash generation.

"Apple Inc. (NASDAQ:AAPL) is a very different company than Anthropic," Damodaran noted. Bears will argue that the gap in revenue and earnings "cannot be explained away with hope and potential."

His core argument is that every pricing multiple hides a discounted cash flow model.

"When you price companies, you're making exactly the same assumptions I will have to make to value the company. You're just making them implicitly," he said.

In his view, to justify $1.5 trillion to $2 trillion, Anthropic needs explosive revenue growth, operating margins of 30% or more, compute costs that grow more slowly than revenue and a risk profile that settles toward average as regulatory threats ease.

As cautious as ever, Damodaran sees the story as "a plausible one," but warns that delivering on it is a whole different matter.