Anthropic has a mega IPO – and an uncertain path ahead

Wait 5 sec.

4 min readNov 20, 2026 06:23 PM IST First published on: Nov 20, 2026 at 06:23 PM ISTOn November 19, Anthropic, the maker of Claude models, listed on Nasdaq in a record IPO that raised $81 billion at a valuation of about $1.5 trillion. While headlines will dwell on the numbers, two developments around the listing deserve attention. First, the timing. Just two weeks earlier, 56 per cent of surveyed voters post the US midterms said data centres were a crucial factor in their vote, with 76 per cent reporting they would oppose a new buildout in their community. Anthropic has in the past committed to absorbing power price hikes and tried to quell apprehensions about environmental damage.Second, in March, the US Department of War designated Anthropic a “supply-chain risk”, a label previously only reserved for foreign adversaries when it refused to blur its red lines, like the deployment of its models for mass domestic surveillance and fully autonomous weapons. The company still carries this label, but that doesn’t seem to have affected its civilian prospects, as its filings indicate. Reportedly, several federal agencies are also working around the ban to test its models. The company, on its part, has since attempted to repair relations with the White House through closed-door diplomatic meetings. In June, it suspended the release of Fable and Mythos worldwide in an attempt to comply with the government’s mandate to block foreign citizens from accessing these models.AdvertisementTaken together, both developments show the same shift in technology geopolitics. Private technology firms and their national governments are becoming more closely aligned. We are now firmly in an era in which technology is treated as a pillar of national power. Governments are willing to step in and reshape private firms’ interests and incentives. Meanwhile, firms are becoming geopolitical actors in their own right, absorbing the state’s mandates and carrying its strategic priorities into global markets.Must Read | The AI bubble has lost its bubble-blower. But don’t grieve just yetAnthropic’s messaging reflects this. In its S-1 prospectus, it devoted roughly 90 pages to “risk factors”, including doomsday warnings that its AI systems could pose “catastrophic or existential risks to humanity.” Its CEO has gone further, likening the export of advanced AI chips to China to “selling nuclear weapons to North Korea”. It has taken a hawkish stance on China, presenting frontier AI as an existential threat and a national-security asset while positioning itself as the industry’s responsible steward is self-serving at best. Given that these firms are so important and closely tied to the US administration, this posture could shape US policy towards export controls and have implications for the rest of the world, especially with the resignation of Jensen Huang, a key voice advocating for the sale of chips to China.The pull towards the state is strongest for AI model companies because of a structural weakness in their business models. Anthropic’s revenue is growing fast, but still dwarfs the roughly $518 billion in compute the company has committed over the next decade. About 80 per cent is non-cancellable and payable regardless of usage. Demand is far less certain, and many of its largest customers, who accounted for nearly a quarter of its 2025 revenue, have no long-term contracts. A firm with fragile financials is likely to find its interests drifting, almost naturally, towards those of the state, which can offer stable contracts, energy and permits, protection from rivals and a regulatory moat.AdvertisementReddy is a researcher at Takshashila Institution, Bengaluru. Views are personal