Trading Strategies & Tech, Markets & Investing
Anthropic’s $2 Trillion IPO Puts AI Infrastructure Financing Under Scrutiny
09 Oct 2026
deVere Group CEO Nigel Green warns that Anthropic’s proposed public listing could test investor confidence in the increasingly interconnected financing model supporting the AI infrastructure boom.
8 October 2026 — Anthropic’s planned Nasdaq listing is becoming a test not only of investor appetite for artificial intelligence, but of the financial structures supporting the wider AI infrastructure boom.
The AI company is reportedly pursuing an IPO that could value it at as much as $2 trillion, according to commentary from deVere Group and recent reporting on the company’s leaked prospectus. The filing shows Anthropic generated approximately $4.6 billion in revenue in 2025 while recording a $42 billion net loss, including a large accounting charge linked to convertible instruments. Its operating loss was approximately $8 billion.
For investors, the valuation raises a central question: how much future AI growth is already being priced into the company before it enters public markets?
Nigel Green, CEO of deVere Group, argues that the answer cannot be separated from the increasingly interconnected flow of capital between AI developers, cloud providers, chipmakers and infrastructure companies.
A $2 trillion valuation demands extraordinary growth
At the proposed valuation, Anthropic would arrive on public markets as one of the most valuable technology companies in the world.
Green argues that such a price assumes an exceptionally favourable growth trajectory.
“A $2 trillion valuation on $4.6 billion of revenue asks investors to pay today for a future where almost everything goes right.”
Independent investment research firm New Constructs has taken an even more sceptical view. Its analysis values Anthropic at around $144 billion under one scenario, more than 90% below the proposed IPO valuation, even while assuming extremely rapid revenue growth and improving profit margins.
That does not mean Anthropic cannot ultimately become enormously valuable. It does illustrate how sensitive current valuation expectations are to future revenue, margins and the scale of AI demand.
Anthropic’s infrastructure commitments are enormous
The company’s capital requirements are another important part of the investment case.
Reuters reported that Anthropic’s leaked prospectus includes approximately $518 billion of cloud, compute and infrastructure obligations.
That scale reflects one of the defining features of frontier AI development: building and operating leading models requires vast amounts of computing capacity, chips, electricity and data-centre infrastructure.
The result is that AI companies are increasingly tied into long-term financial relationships with the companies supplying that infrastructure.
Green argues that this has created what he describes as a “circular” system in which money can move between the same groups of investors, infrastructure providers and AI customers.
His concern is not that every transaction is artificial. Rather, it is that financial interdependence can make it harder for investors to determine how much demand is ultimately independent and how much is reinforced by capital moving around the same ecosystem.
The Bank of England is watching the same risk
Green’s warning echoes concerns raised by the Bank of England’s Financial Policy Committee.
In September, the Bank said AI-related financing was expanding rapidly and specifically warned that increasing indebtedness, opacity and, at times, “circular arrangements” could make risks more difficult to assess and amplify losses if investor expectations disappoint.
The Bank said global AI-related debt issuance had reached approximately $450 billion by early September 2026, more than double the total issued during all of 2025.
Its July Financial Stability Report had already highlighted the growing role of debt in AI infrastructure, noting that hyperscaler borrowing was expanding across investment-grade credit, leveraged finance, structured products and private markets.
The issue therefore extends well beyond Anthropic.
AI has become a capital-markets story as much as a technology story.
Debt is becoming more important to the AI buildout
The first phase of the AI boom was supported heavily by the cash flows and balance sheets of some of the world’s largest technology companies.
That is beginning to change.
The Bank of England says AI infrastructure increasingly depends on outside financing, including corporate bonds, private credit and structured financing arrangements. It also warned that declining free cash flow among major hyperscalers could increase dependence on favourable refinancing conditions.
This matters because the economics of AI infrastructure depend heavily on assumptions about future demand.
If usage continues growing rapidly and AI produces major productivity gains, the investment may prove justified.
If demand, pricing or margins disappoint, the same leverage that accelerated expansion could magnify losses.
Circular financing does not mean demand is fake
The distinction is important.
The existence of interconnected financing arrangements does not mean AI demand is fabricated.
Anthropic’s revenue expanded dramatically in 2025, and enterprise adoption of AI continues to grow. Reuters reported that revenue increased roughly twelvefold during the year.
The question for traders and investors is instead whether infrastructure spending, valuations and financing have moved ahead of sustainable cash generation.
Green puts the issue this way:
“Some of this demand is genuine and powerful. The question nobody can yet answer with confidence is how much of it is the same dollars doing laps.”
For public-market investors, that distinction becomes more important as private AI companies begin listing on stock exchanges.
Public investors inherit the valuation risk
An IPO changes who ultimately carries the financial exposure.
Private investors can realise gains through a public listing, while index funds, pension funds and retail portfolios may begin acquiring the stock after it enters major benchmarks.
Green warns that an IPO at this scale could distribute AI valuation risk much more widely.
That does not make an IPO inherently problematic. Public markets routinely finance growth companies with uncertain future earnings.
But at a valuation approaching $2 trillion, the margin for disappointment becomes narrower.
The company would need to deliver extraordinary long-term growth to justify such expectations.
Echoes of previous investment cycles
Green compares the current structure with the telecommunications boom around the turn of the millennium.
Telecom equipment suppliers sometimes financed their customers, enabling those customers to continue buying equipment and supporting reported revenue growth.
The technology itself ultimately transformed the global economy.
Many investors still lost money.
The historical comparison is useful because technological importance and investment returns are not the same thing.
A sector can reshape society while individual companies, valuations or financing structures prove unsustainable.
The real test for Anthropic investors
Anthropic’s IPO therefore poses two separate questions.
The first is technological: can Claude and Anthropic’s broader AI platform become foundational infrastructure for businesses and consumers?
The second is financial: what price should investors pay today for that possibility?
New Constructs argues that the proposed valuation leaves little room for anything short of exceptional execution. deVere’s concern is broader: that investors should examine not only Anthropic’s growth, but the financial relationships underpinning the AI infrastructure ecosystem around it.
As more AI investment is funded through debt and public markets, those questions are becoming increasingly relevant to traders far beyond the technology sector.
AI may still prove to be one of the most important technological shifts of the century.
But as Green puts it:
“Brilliant technology and a sound valuation are two very different things.”
About deVere Group
deVere Group is an independent financial advisory organisation serving international, mass-affluent and high-net-worth clients.
According to the company, it has more than 80,000 clients globally and approximately $14 billion under advisement.
