OpenAI and Anthropic are targeting IPOs within the next 12 months, each seeking valuations between $1.5 trillion and $2 trillion—projections built on assumptions of sustained technological dominance and explosive revenue growth.
AnthropicRev run rate is estimated at $65 billion this year, up sevenfold from 2025. Yet this calculus depends entirely on maintaining a performance and cost advantage that is rapidly eroding.
Chinese open-source and open-weight AI models are closing the capability gap while undercutting U.S. offerings on price. DeepSeek and Kimi now deliver comparable results for business, government, and military applications—in some cases at less than 50 percent of the per-token cost of proprietary American alternatives. Enterprise customers are noticing. Adoption of Chinese models among U.S. companies is accelerating precisely because the performance delta no longer justifies the price premium.
Nvidia CEO Jensen Huang recently acknowledged that Chinese AI models are "excellent," a candid assessment that undercuts the narrative of unchallenged American superiority. Yet U.S. policymakers are moving to restrict their deployment, citing espionage concerns. This policy tug-of-war leaves the market uncertain: if Chinese models are genuinely dangerous, broader restrictions could follow; if they are merely competitive, the restrictions may prove unenforceable or temporary.
A second threat cuts dee capital intensity. OpenAI, Anthropic, and their peers require hundreds of billions in data center spending to sustain their competitive position. That spending makes sense only if they can maintain pricing power and market share. If Chinese alternatives prove adequate for most use cases, the return on those infrastructure investments collapses—and with it, the revenue projections underpinning both companies' valuations.
Data center buildouts in the U.S. are already slowing. Permitting delays and project cancellations have cascaded through the supply chain, raising questions about whether capital deployment will match the sector's bullish forecasts.
For IPO investors, the core bet has shifted. Valuation multiples of $1.5 trillion to $2 trillion assume American AI firms remain years ahead of competitors and can sustain premium pricing. Chinese competition and falling barriers to entry make that assumption increasingly fragile. Watch whether OpenAI and Anthropic narrow their IPO valuation targets before launch—a tell-tale sign of pressure from early backers and underwriters. If they do not, watch how the IPO pricing performs relative to comparable high-growth software names. A significant valuation discount versus historical comparables would confirm that the market has repriced the competitive threat.
