Anthropic has filed paperwork for an initial public offering, signaling its intent to list on public markets. The AI research company submitted its S-1 registration statement on Monday, revealing that it lost about $42 billion in 2025 and generated $4.6 billion in revenue. The filing also confirms the company plans to trade under the ticker symbol ANT.
What the Filing Shows
The prospectus offers a detailed look at Anthropic’s financial health. Venture capitalist Sarah Guo highlighted the shift in operating profit in her initial reaction to the filing. She also pointed out a mismatch between the company’s consumption-based revenue model and its long-term obligations for computing power.
Guo expressed curiosity about how Anthropic will operate under the scrutiny of quarterly public reporting and how public perception might shift once individual investors own a piece of the company. Gary Marcus, a well-known critic in the AI space, raised concerns about the absence of a full third-quarter update, arguing that the company omitted a full Q3 update and was “trying to skirt the truth.”
The S-1 document reveals the sheer scale of Anthropic’s growth ambitions. Harrison Rolfes noted that while the company is expanding rapidly, it is doing so at a high cost. The filing shows $11.6 billion in revenue for the second quarter, but also exposes massive financial commitments that have not yet begun. Anthropic has $321.4 billion in future lease payments and $265.6 billion in fixed minimum purchase commitments.
Rolfes explained that these obligations are tied to financing arrangements with Google, Amazon, and Broadcom, meaning the suppliers are also lenders. He emphasized that the price range, third-quarter revenue figures, and total compute commitments are still blank in the current version of the filing, suggesting the company will need to amend the document soon.
Despite the heavy financial commitments, some analysts see efficiency in Anthropic’s model. SemiAnalysis highlighted that the company is “profitable on an operating income basis as of the June quarter.” The research firm called this the “most efficient scaling of a business in the history of capitalism,” pointing to the company’s raised capital compared to its cash and investments.
Read Also: Ex-Netflix Employee Turns Japanese Kindergarten into Cinema
Anthropic’s founders retain significant voting power, a fact that Rolfes noted will prevent public shareholders from controlling the company. Sam Ro, a financial journalist, argued that the company’s long section on AI risks should not worry investors. If AI creates an existential catastrophe, Ro suggested, having money might not be the primary concern.
Dan Shipper celebrated that the S-1 does not open with a warning that the AI might kill or unemploy everyone. Dadisi Olutosin asked a critical question for potential buyers: when does an extraordinary company become an overpriced investment?
At a reported valuation of $1.5 trillion to $2 trillion, buyers would pay about 101 times the company’s trailing revenue, compared with roughly 20.5 times the revenue pace at which Series H investors bought in. Innovation can transform an industry, Olutosin wrote, but that does not make every entry price attractive.
Nick Moran argued that the $80.7 billion loss in the first half of 2026 was largely an accounting story, noting that Anthropic became more valuable while reporting the loss. He said the bigger question for investors is the price they will pay. While revenue growth remained “spectacular,” Moran wrote that it slowed from the second quarter to the third as the company sought a higher valuation.
Investor Reactions and Valuation Debate
Several analysts noted that the filing comes as investors become wary of debt-fueled projects. The cost of developing frontier models remains high. Companies are turning to public markets for fresh investments to stay competitive. The stock price may rise or fall after it opens on the market.
