Anthropic Is Valued at $965 Billion Privately and Could List at $2 Trillion
Anthropic's latest private valuation is $965 billion, and the company is expected to list publicly this year at a valuation of as much as $2 trillion, according to TechCrunch.
The early investors have seen their positions compound sharply. Eric Schmidt invested in Anthropic's Series A in 2021, when the startup was valued at $461 million. He owns about 1% of the company, a stake worth nearly $10 billion at the current private valuation. Iconiq put $6 billion into the company last year, when it was valued at $183 billion. That stake is now worth $17 billion. Menlo Ventures led the Series D and has invested more than $1 billion over time. Its stake is now worth $14 billion.
Sequoia invested $3 billion earlier this year, when the company was valued at $380 billion, for about 1% ownership. Spark Capital's path is the steepest of the group. It invested $75 million in the 2023 Series C round at a $5 billion valuation, and that position has grown more than 90-fold to $7 billion. Spark is the only venture capital firm known to hold a board seat at Anthropic. Sound Ventures, launched by Guy Oseary and Ashton Kutcher, invested $35 million in the same Series C round and later added $400 million.
The valuation path is steep in its own right. Each funding stage is called a Series round, labelled by letter in order. Anthropic completed a $13 billion Series F led by ICONIQ in September 2025 at a $183 billion post-money valuation, meaning the value once the new money is counted. In February 2026 it announced a $30 billion Series G at a $380 billion post-money valuation, led by GIC and Coatue. In May it raised $65 billion in a Series H led by Altimeter Capital, Dragoneer, Greenoaks, and Sequoia Capital. Earlier in the year, TechCrunch reported that the company was targeting a valuation of about $900 billion, and that buyers had indicated they had $2 billion in cash ready to deploy. Anthropic said on June 1 that it had confidentially submitted a draft S-1, the registration document filed with the SEC before a public listing.
The company's operating figures have also been disclosed in part. Reuters reported third-quarter revenue of $16.8 billion, up 45% from the previous quarter. The same outlet, citing Anthropic's IPO prospectus, reported that the company lost $42 billion in 2025 and spent $7.33 billion on compute, the processing capacity that trains and runs AI models. Anthropic has also said that roughly 20% of its shares will be dedicated to the public good, through a public benefit fund it described in an announcement on October 18.
The broader context here is the gap between the revenue line and the loss line. A company growing revenue at the reported rate, while carrying compute costs at the scale reported for 2025, will be valued by public-market investors on forward assumptions about margin, the share of each revenue dollar left after costs, not on trailing figures alone. The step from a private mark of $965 billion to a public target of up to $2 trillion is the assumption that the market will underwrite that trajectory. Private marks are set in funding rounds and secondary sales, and they carry different disclosure and liquidity terms than a listed share price.
For the backers, the arithmetic is simpler. Ownership percentages are fixed at a given round, so the value of each stake moves with the valuation alone. Schmidt's roughly 1% at $965 billion is nearly $10 billion, and Spark's position, bought at a fraction of the current mark, has grown with each step up the ladder. A public listing would convert those paper gains into tradable shares, subject to lockups that restrict when insiders can sell and the disclosure obligations that come with a registered offering.
In my view, the more consequential question for the sector is what a public Anthropic makes possible for the infrastructure that supports it. Compute spending is the largest line in the company's reported cost base, and a listed balance sheet would give it a different way to finance that buildout. Whether that capital ends up accelerating model capability, inference efficiency, or the next generation of data-center commitments is something the prospectus and the first earnings calls will begin to answer. For now the company is growing fast, spending heavily, and preparing to ask public markets to price the difference.
