
The largest IPO in history
The high-profile stock market debut of the company behind the AI assistant Claude, followed by a share-price correction the very next day, illustrates both the promises and the uncertainties surrounding the business model of artificial intelligence. Given the growing opposition to the establishment of data centers, the future rollout of AI will also depend on what the public is willing to accept.
Artificial intelligence (AI) has less and less room to disappoint. Even as investors question the technology’s promised profitability, Anthropic, the developer of the AI assistant Claude, carried out the largest initial public offering (IPO) in history on Thursday, November 19, raising $81 billion. After climbing 17% on its first day of trading, the stock fell 8.5% the following day, amid a broader market rout that saw the Nasdaq drop 5.3%, wiping out all its gains since January 1.
Coming five months after SpaceX’s arrival on Wall Street and ahead of a potential IPO for OpenAI, this move adds a new dimension to American capitalism’s massive bet on AI. Yet, behind this display of strength, new vulnerabilities are emerging.
The sums raised by Anthropic may seem extravagant, but the company’s financial requirements are equally vast. Now sitting on a cash pile of $230 billion, the company plans to spend $900 billion on its IT infrastructure by 2045. As spectacular as its revenue growth may be, it will need to remain robust for years to come to make such investments profitable, a formidable challenge.
The development of AI
The development of AI is also characterized by the growing interdependence of the players involved.
Nvidia, Microsoft, and Google, all major suppliers to Anthropic, are also among its financial backers. From semiconductor manufacturers and electricity providers to banks and investment funds, entire sectors of the economy are pinning their growth prospects on the same set of assumptions. Should those assumptions be called into question, it could trigger a chain reaction, impacting everything from industrial investment to the valuation of global savings.
The strength of the U.S. economy already owes much to AI-related investments and the wealth effects driven by the surge in tech stocks. However, this dynamic is accompanied by upward pressure on the prices of energy and certain types of equipment, while the cost of financing makes projects with uncertain profitability more vulnerable.
Friday’s stock market correction offers a glimpse of these risks. Investors have priced in the potential gains from AI to such an extent that even a slight gap between their expectations and the actual revenue generated could trigger a reversal.
There were even billboards in Times Square displaying the branding of the company Anthropic on the day of its initial public offering in New York, November 19, 2026. Investors must also contend with growing public opposition to the siting of data centers, a sentiment highlighted during the midterm election campaigns. The pace at which this infrastructure is deployed will ultimately depend on what local communities are willing to accept.
Some industry players themselves acknowledge the need to set limits on this expansion. Anthropic co-founder Dario Amodei has pledged to resist both commercial and political pressures, even if it means slowing the development of his company’s models for the sake of safety. Transparency requirements for publicly traded companies could facilitate better oversight of the sector. But how much leeway will executives have to rein in the race for ever-more-powerful models if economic prosperity—and the value of people’s savings—become increasingly dependent on them?
The danger is that AI could become so essential to economic prosperity that public authorities lose their ability to regulate it, both in terms of its technological development and its environmental impact.

