Anthropic's path to a public listing has gone from filing paperwork to naming a day. The company's Form S-1, lodged with the US Securities and Exchange Commission on 26 October, confirmed what had been rumoured for months: a listing on the Nasdaq Global Select Market under the ticker ANT. What has since come into focus is the timing, and it is tight. Market reporting now points to shares beginning to trade this Thursday, with an IPO calendar listing 19 November as the expected first trade date, contingent on the SEC's registration becoming effective in time.
The roadshow that preceded this week put a number on the ambition, if not yet a settled one. Anthropic is offering 825 million shares in a price range of $70 to $78, an indicative band rather than a locked-in figure, with the final price to be set once order books close. At the top of that range, the implied valuation runs well past a trillion dollars; even conservative readings of the roadshow numbers put the company among the largest technology listings of the year by any measure.
That scale has not translated into investor consensus. Roadshow attendees have reportedly argued over whether Anthropic is worth something closer to $1.2 trillion or nearer $2 trillion for fundamentally the same business — a spread wide enough to swallow the entire market capitalisation of most national stock exchanges. The disagreement says less about Anthropic specifically than about how unresolved the market's model for valuing frontier AI labs still is: revenue multiples that would look absurd applied to almost any other industry are, in this one, treated as a live debate rather than a red flag.
Anthropic's own numbers give both sides of that debate something to point to. The company's first-half 2026 revenue has been reported in the $16.2-to-$16.3-billion range depending on the source, a growth trajectory that would be extraordinary for a company barely years into commercial operation. Set against that is a financial picture dominated by the enormous cost of training and running frontier AI models, a scale of spending that has made profitability, by Anthropic's own account, a question for later years rather than this one — the same bet every major AI lab is currently asking public and private investors alike to make.
Beyond the United States, Anthropic has already tested retail appetite more directly: an Australian retail offering opened ahead of the Nasdaq listing, giving smaller investors there a route into the stock before the main US debut even priced. That sequencing, usually reserved for companies confident enough in demand to let it build gradually across markets, suggests Anthropic's bankers are betting the valuation argument resolves in their favour once trading actually begins.
Whichever number the market lands on once shares start changing hands, Thursday's debut will function as a referendum on more than one company. Anthropic going public at any price north of a trillion dollars would cement the idea that the AI labs racing against OpenAI are not merely well-funded private ventures but public-market-scale businesses in their own right — a status that invites exactly the quarterly scrutiny and short-selling pressure that privately held AI labs have so far been spared.