Anthropic IPO: From $47B Run Rate to $200B 2028 Growth Bet

For founders and venture investors, Anthropic's IPO is a pivotal moment in how late-stage AI companies are priced. A $190 billion to $200 billion 2028 forecast from a $47 billion run rate extends the long-horizon revenue underwriting used by Cerebras and SpaceX into the public markets.

Anthropic is asking public-market investors to underwrite a revenue trajectory that stretches further into the future than most IPO valuations do. According to two people familiar with the company's financials, the AI developer is projecting 2028 revenue of roughly $190 billion to $200 billion, a figure not previously reported. Bankers and investors are using enterprise value-to-revenue multiples based on those long-dated forecasts, four sources said, rather than anchoring valuation primarily on current revenue or near-term expectations. The gap between that 2028 projection and the $47 billion revenue run rate Anthropic publicized as recently as May 2026 is enormous: it implies roughly a fourfold expansion in about two years. That is the scale of growth being underwritten in what could be one of the largest IPOs on record.

“

A $190 billion to $200 billion 2028 forecast from a $47 billion run rate extends the long-horizon revenue underwriting used by Cerebras and SpaceX into the public markets.

Valuing a high-growth software company on a revenue multiple is not new, but extending the multiple two years out is less typical. The approach reflects the speed at which Anthropic's business is expanding and the difficulty of setting benchmarks for a company still spending heavily to build out its AI infrastructure. The sources said that the pace of AI investment has already caused pullbacks in many of the most popular tech stocks in recent months, including firms viewed as comparable to Anthropic. That backdrop makes the two-year valuation basis a double-edged sword: it highlights the potential size of the business while asking public markets to price in revenue that does not yet exist and may depend on continued extraordinary capital deployment.

The forward-looking method has recent precedents among the fastest-growing companies to reach public markets. Backers of Cerebras Systems cited 2028 revenue expectations in the runup to its IPO this year, and SpaceX projections extended as far as 2029 before the company went public at a record valuation in June. Those cases suggest an emerging playbook for frontier technology listings: when current earnings are thin and growth is nonlinear, underwriters and investors look to a future year in which scale may translate into profitability. But the playbook also concentrates risk. If forecasts disappoint, the valuation reset can be severe because the original multiple was applied to a much larger revenue base far in the future.

At the center of the Anthropic valuation is the question of whether revenue will rise faster than the costs required to support that growth. The company is still spending heavily on computing power, model training, and hiring, which pressures margins and makes current profitability an unreliable basis for valuation. Investors are betting that as Anthropic scales, the ratio of revenue to infrastructure and personnel costs will improve, allowing margins to expand. That is a reasonable assumption for software businesses with operating leverage, but it is far from guaranteed for AI companies whose products require unusually expensive compute and who must keep paying for frontier model development to remain competitive.

The IPO's outcome will have broad market implications. A successful listing at a valuation consistent with $190 billion to $200 billion in 2028 revenue would validate a new, more patient underwriting model for AI and could encourage other private AI leaders to pursue public exits despite thin current profits. It would also signal that public investors are willing to absorb the capital intensity of AI in exchange for future growth. Conversely, if AI spending concerns and tech-sector pullbacks intensify, the same forward multiples may compress, lowering the valuation that investors are willing to accept. The sources' framing of the valuation as hinging on a 2028 forecast underscores the fragility.