Anthropic's confidentially-filed IPO is more of a timestamp:

From my post on LinkedIn:
Everybody in AI understands the same thing. There is a window, it is open, and nobody knows how long it stays that way. Capital is abundant, public-market appetite is back, the AI narrative still has room to suspend disbelief, and private valuations have become large enough that the only natural next buyer is the public market.
This is the whole AI stack racing to convert narrative into cash before the narrative closes. The winners need money for compute. The almost-winners need money to prove they are winners. The infrastructure players need demand to justify capex, and the public markets, for now, are still willing to underwrite the idea that intelligence-as-a-service will be one of the great business models of the next decade.
That is the strange beauty of IPO windows. They look obvious only while they are open. Six months later, the same company, the same revenue curve, and the same burn rate can be read very differently. Anthropic knows this and so does everyone else.
My advice: avoid buying the story at the IPO price. Buy only after the market has killed the story once.
Here’s why:
The IPO is the seller’s moment, not the buyer’s moment.
Founders, insiders, late-stage investors and bankers choose the listing window. They do not choose it because public investors are getting a bargain. They choose it because the narrative is strong, liquidity is available, and the valuation is high enough to sell.The last 5 or 6 years have shown the same pattern again and again.
Many IPOs came to market with heroic valuations, perfect stories and huge expectations. Then, 6 to 12 months later, the market did what markets do: it stripped away the perfume. Growth slowed. Margins mattered. Lock-ups expired. Insiders sold. And the real price appeared.

Source: Financial Times, Goldman Sachs
…the same story is for European IPOs:

Bloomberg
The first price is marketing. The later price is discovery.
At IPO, investors buy the trailer from the S-1 form. Usually, this sounds bombastic… “AI, space, Mars, agents, compute, cloud, sovereignty, disruption”. After a few quarters, they start watching the actual movie: revenue quality, cash burn, gross margins, dilution, governance, customer concentration, and free cash flow.Anthropic may be an extraordinary company, but the valuation risk is enormous.
The company sits at the center of the AI boom, which means the IPO price will likely include not only the business as it exists today, but also the dream of what it might become. That is dangerous. When perfection is priced in, even very good execution can disappoint.The core question for Anthropic is brutally simple: “Who captures the margin?”
Anthropic sells intelligence, but it must buy massive amounts of compute. If AI models become more commoditized over time, pricing power may shift to cloud providers, chipmakers, enterprise distribution, or workflow platforms. Public investors may think they are buying a software-margin business, only to discover something closer to an infrastructure-heavy business.SpaceX is even more dangerous because the story is almost too good.
SpaceX combines rockets, satellites, Starlink, defense, Mars, AI, and Elon Musk into one of the most seductive narratives in modern capitalism. That is exactly the problem. When a story is this powerful, the valuation can easily move from analysis to faith.Starlink is the real business..for now. The rest is optionality.
SpaceX may deserve a premium because Starlink is a rare asset with global infrastructure, recurring revenue, strategic importance, and huge scale. But investors must separate the cash-generating business from the bundle of dreams around Mars, AI, launch dominance, and future space infrastructure. Optionality is valuable, but at IPO, it is often sold at full price.The lock-up expiry is the first real stress test.
In the first months after an IPO, supply is artificially limited. Later, employees, early investors and insiders may be allowed to sell. That is when the market learns whether demand is truly deep, or whether the IPO was simply a scarcity event wrapped in a great narrative.The right strategy is “wait”
Wait for the first earnings reports. Wait for the first disappointment. Wait for the first lock-up expiry. Wait for analysts to stop selling the dream and start debating the numbers. With Anthropic, watch compute costs, gross margins, enterprise adoption, burn rate and dilution. With SpaceX, watch Starlink profitability, Colossus renting (revenue), capex, launch economics, AI losses, government exposure and governance.
Concluding, here is Warren Buffett:
The stock market is a device for transferring money from the impatient to the patient.
Have a nice day!
Sorin

WIMM for companies: If you would like to offer the newsletter as a benefit to other employees in your company, please contact me directly. ([email protected])
Work with me: If you are interested in working with me on your personal portfolio, again, contact me. I’m a certified financial consultant.
Finally, “Where is my moat?” is designed for individual readers, though the occasional forward is absolutely fine.
Thanks for your support & have a wonderful day!

