Anthropic IPO filing tests $965bn AI valuation

Anthropic’s $965bn private valuation moved closer to Wall Street on Monday after the Claude maker confidentially filed IPO papers with the Securities and Exchange Commission, Reuters reported. For public investors, the filing creates the first formal test of a price tag built in the private AI boom.

Confidential filing is still process, not a priced deal. It begins an SEC review that lets Anthropic keep financial details private while management gauges whether market conditions can support a share sale. In a statement cited by CNBC’s report on the filing, the company said any offering would depend on those conditions and other factors. Regulators get the first look; bankers get the window later.

“This gives us the option to go public after the SEC completes its review,”
Anthropic, according to CNBC

A valuation story has become a listing event. Anthropic announced a $65bn Series H round last week that valued the company at $965bn post-money, up from $380bn in February, CNBC reported. At that size, even a small free float would put the deal near the centre of the equity-capital-markets calendar.

OpenAI is the obvious comparison, but the race has changed. Model rivalry now sits beside timing, disclosure and the price public funds are willing to pay. CNBC’s figures put OpenAI at an $852bn valuation in late March, and Reuters said the two companies are moving toward possible Wall Street debuts. Bankers have a blunt question to answer: will public buyers accept private-market prices already close to $1tn?

Valuation meets the SEC

Through confidential review, Anthropic can revise its draft registration statement with SEC staff, wait for a stronger market window and hold back public disclosure until closer to a roadshow. Investors still lack the figures they will need: revenue quality, cloud-computing commitments, gross margins, customer concentration and the capital spending required to train and serve frontier models.

Available numbers frame the debate anyway. Anthropic’s reported $47bn revenue run rate implies a price-to-run-rate-sales multiple of about 20 times at the $965bn valuation. That is not a normal software multiple. Buyers would be underwriting demand for Claude, enterprise adoption and pricing power that keep outrunning infrastructure costs.

Valuation risk is why the IPO may test more than appetite for AI exposure. University of Florida IPO specialist Jay Ritter told Fast Company that at the potential prices being reported, it would be hard for an investor to come out ahead over three years. Scale is not the issue. Whether that scale has already been capitalised is.

“At the potential prices that have been reported, it would be very difficult for an investor to come out ahead in a three-year period,”
Jay Ritter, quoted by Fast Company

Crowded IPO window

Anthropic is moving into a listing calendar that could be unusually heavy with megacap private technology companies. Bloomberg has reported that SpaceX is targeting an IPO valuation of at least $1.8tn, while other coverage has put OpenAI on a parallel track. The Guardian framed the race as a wider contest among AI-linked groups for capital, talent and public legitimacy.

Market capacity for giant growth listings is not unlimited. A run of AI and space deals would compete for the same long-only funds, sovereign wealth funds and growth portfolios. Strong early trading in one debut could help the next. A weak print, or a quick drawdown after pricing, would make every near-$1tn valuation harder to defend.

For now, Anthropic has the clearest capital-markets milestone. Its paperwork is in, the SEC review is the next gate and the valuation reference point is fresh. The open question is whether investors treat Claude’s growth as a scarce public AI asset, or as a private-market mark that reached Wall Street before the cash-flow evidence did.

Bankers can sell a simple story. Executing it will be harder. Anthropic needs enough disclosure to persuade public investors that its revenue run rate is durable, without giving them a reason to focus more on compute costs or customer concentration than on growth. That balance will decide whether the filing becomes the AI boom’s flagship IPO or a reminder that private rounds do not set public-market clearing prices.