Anthropic Is Set for the Largest IPO in History—the Hard Part Is Pricing It
FT says Anthropic's IPO is shaping up to be the largest in history. Unhedged pushes back on the high-flying narrative and discusses a more practical question: how should this AI company actually be valued? In the video, the host speaks into a microphone and says there is something absurd about this.
FT's tweet says it in one line: Anthropic's IPO is shaping up to be the largest in history.
Then FT's Unhedged podcast pulls the focus back to valuation. Rob Armstrong and John Foley do not take flight on the word "largest." They unpack a messier question: why is this company so hard to price? In the video, a host sits in front of a microphone as the subtitles flash "One of the absurd things about this." The sense of absurdity comes from the contrast. The larger the IPO, the more the public market needs a persuasive valuation language—and AI companies are precisely the ones most lacking stable benchmarks.
Anthropic is the developer of the Claude model family. The revenue curve for a generative AI company can be steep, and so can the cost curve. Compute procurement, cloud service contracts, talent compensation, model training, and inference expenses all pull the income statement into a jagged line. Traditional P/E and P/S ratios, applied here, easily become after-the-fact explanations. Even harder is the pace of competition. Model iterations are measured in months; a capability that leads today may become table stakes next quarter. Valuation has to judge a river that is still changing course, and the product is only the surface.
So it is not surprising that "largest IPO in history" and "hardest to value" appear side by side. IPO pricing requires translating a story into numbers. What Anthropic has to face is whether the public market is willing to accept a new valuation framework. The original FT link currently triggers a security check; what can be confirmed is that the episode's theme centers on this difficult problem.
One reply under the tweet is very direct: how to get in early with the institutions and then dump everything later the same day. This joke captures the other side of a large IPO. Institutional pricing, retail investors taking the bag, first-day volatility—these old problems get magnified in AI companies. The larger the scale, the more pricing resembles a public exam. It tests who is willing to pay how much for uncertainty, and it also tests the market's patience with the AI narrative.
FT original article: https://ft.trib.al/vStCmGl
发布时间: 2026-10-29 07:26