Anthropic's planned $2 trillion stock debut faces doubts and a delay

Anthropic's planned stock market debut has been pushed from October to November, and a guest writer at Crunchbase News argues the delay does not fix the real problem. Anurag Gurtu, CEO of AI company Airrived, says the proposed $2 trillion valuation is too high, even though the company is growing fast. This is his opinion, not settled fact.

The strong numbers

Anthropic's third-quarter 2026 revenue was reportedly $16.8 billion, up 45% from the quarter before. That puts its yearly pace at roughly $67 billion. Gurtu does not dispute that. His point is that growth alone does not justify the price.

The warning signs he points to

• NYC Comptroller Mark Levine wrote to CEO Dario Amodei urging a pause, citing risks Anthropic's own filing lists.
• OpenAI says it will not go public in 2026. CEO Sam Altman called an IPO now "ill-advised."
• A $2 trillion chip-stock selloff last week was driven partly by worries about AI building costs.
• A quarter of Anthropic's 2025 revenue came from just two unnamed customers.

The filing's own language includes "existential risks to humanity" from increasingly autonomous AI systems. Gurtu notes that Altman's remark is partly strategic, since it makes OpenAI look responsible, but still calls it a real signal.

The price problem

Gurtu says a $2 trillion value assumes that top-tier, closed AI models can keep charging more than free alternatives. He says that edge is fading. Open-weight models are ones whose files anyone can download and run on their own. He cites enterprise use of them rising from 23% to 67% in 2026, with companies reporting 70% to 90% savings.

One example is DeepSeek R1, priced at 7 cents per million input tokens, the small chunks of text AI services bill by. An independent analysis puts Anthropic's comparable input and output prices at roughly 18 times and 23 times higher. Gurtu adds that the capability gap with closed models reportedly closes within six months of a new release.

He also cites research firm New Constructs, which says the valuation needs a 10% after-tax operating margin held for a decade and $358 billion in yearly after-tax operating profit by 2035. He says the price equals 435 times 2025 revenue, set against $518 billion in infrastructure commitments.

The wider picture

Gartner forecasts $2.59 trillion in worldwide AI spending in 2026, up 47%. It also sees at most $3.6 trillion in revenue tied to AI data centers through 2030, against $8.5 trillion needed to cover planned costs and a 12% return. An MIT NANDA report found 95% of organizations examined were getting no return from generative AI. KPMG says 7% of surveyed leaders have established a return.

What to keep in mind

Gurtu sells enterprise AI tools, so he has a stake in arguing that the model is not the product. His view is that the winners will be companies that make any model work reliably inside a business. For anyone holding AI stocks or index funds, the debate is a reminder that a great product and a fair price are different questions. Anthropic has not been quoted in this material.