Anthropic's shares are set to start trading on the Nasdaq on Thursday at $83 each, a price that has pushed the company's value as high as $1.5tn (£1.14tn). The artificial intelligence company behind the chatbot Claude would become one of the most valuable firms in the world, and the first US company focused exclusively on AI to go public. Enthusiasm is easy to understand. Demand is reportedly more than 10 times the number of shares available, according to one early investor who spoke to the BBC on condition of anonymity.

Enthusiasm should not be confused with proof of a business. Yang Wang of Counterpoint Research put the central issue plainly: whether Anthropic's business "can keep pace" as hundreds of billions of dollars in financial obligations come due over the next couple of years. The company expects to raise more than $68bn from its initial investors plus another $26bn from private placements, a total of $94bn. That is a great deal of money to raise against a revenue model that the report itself describes as still uncertain.

The wider industry deserves the same scrutiny. The report notes that AI businesses depend on expensive computer chips that are in short supply, on controversial data centres, and on deals with other tech companies. The Federal Reserve has voiced concern about rapid growth in AI-related debt financed through "complex and opaque" structures, and its October minutes said AI-related exposures were becoming increasingly concentrated among a small number of firms. When a single listing is treated as a referendum on a whole sector, that concentration is the risk worth watching.

Anthropic's stated commitment to safety deserves credit, and its refusal to let the US Department of Defense use its tools in an essentially unrestricted way is a notable position. Its CEO, Dario Amodei, has warned that AI tools may soon be powerful enough to threaten human existence. But a safety-focused brand does not remove the need for investors to ask hard questions about cash flow. Public markets can supply the money the company says it needs, and Marta Norton of Empower is right that the listing offers a new window into what AI demand looks like.

What should happen next is simple. Investors should judge Anthropic on whether its revenue can carry its obligations, not on the excitement of its opening price, and regulators should keep a close eye on the debt structures the Federal Reserve has flagged. If Thursday's trading confirms the demand, the company will have proven something real about the market. The harder test comes after the first-day headlines fade.