Anthropic Pursues IPO at Up to $2 Trillion Valuation in Test of AI Pricing
Summary
- Anthropic is pursuing a U.S. stock market listing this year with a target valuation of as much as $2 trillion and plans to raise up to $100 billion in its IPO.
- Last year's revenue was $4.6 billion, about 12 times the previous year, but the company still faced pressure from an $8.06 billion operating loss, a $42 billion net loss, $7.33 billion in compute and infrastructure costs, and $518 billion in contractual obligations.
- Hana Securities said the durability of revenue growth and compute cost efficiency are the key variables in justifying the valuation, and that the Anthropic IPO could become a benchmark for assessing frontier AI company valuations and the infrastructure investment cycle.
Forecast Trend Report by Period



Anthropic is pursuing a U.S. stock market listing that would value the artificial intelligence company at as much as $2 trillion. Despite explosive revenue growth, the company continues to post steep losses and bear heavy compute costs, setting up a test of how public markets will value frontier AI companies.
Bloomberg and other media outlets reported on October 1 that Anthropic is considering starting its initial public offering roadshow in the week beginning November 9 and launching trading before the U.S. Thanksgiving holiday on November 26. Even if the schedule is delayed, the company could still complete the listing this year.
Investors have discussed a valuation of $1.8 trillion to $2 trillion. Hana Securities estimated the IPO could raise as much as $100 billion. If realized, that would surpass SpaceX's $75 billion fundraising in June and could make it the largest IPO on record.
Rapid growth is underpinning the valuation. Anthropic posted $4.6 billion in revenue last year, about 12 times the previous year's level. Its operating loss widened to $8.06 billion from $2.98 billion a year earlier.
Net loss totaled about $42 billion. Of that, roughly $34 billion was a non-cash charge stemming from an increase in the value of previously issued stock-convertible financial instruments.
Compute costs for training and operating AI models also remain a burden. Of the company's $12.65 billion in total operating expenses last year, compute and infrastructure costs accounted for $7.33 billion, or more than half. Contractual obligations tied to cloud, compute and infrastructure spending over the coming years also total $518 billion.
Revenue concentration is another variable. About 25% of last year's revenue came from two customers. Because many major clients are not under long-term contracts, lower usage or contract termination could lead to volatility in results.
Kim Jae-im, an analyst at Hana Securities, said the durability of revenue growth and improved efficiency in compute spending are the key variables in justifying Anthropic's valuation.
Anthropic also plans to maintain a founder-centered governance structure after the listing. Under that arrangement, a separate group made up of seven co-founders would hold 50.1% of the voting rights over major decisions.
Kim said Anthropic's IPO would be the first major case in which public markets fully test the business performance and profitability of a large frontier AI company. The market's assessment after the listing could become a benchmark for judging valuations, disclosure standards and the infrastructure investment cycle for AI companies including OpenAI.