Anthropic's expected debut on the Nasdaq on Thursday should terrify, not excite, investors who care about capital discipline. A company that has accumulated a net loss of $104 billion through the first nine months of 2026 is about to raise over $90 billion more—ostensibly to fund outstanding spending commitments of over $900 billion. This is not visionary capitalism. This is a collective delusion masquerading as progress.

Consider the commitments Anthropic has made: at least $110 billion to Amazon over a decade, $400 billion in cloud capacity obligations, $300 billion to data center and chip partners, and $125 billion to $150 billion for AI chips from Google with delivery starting in 2028. These are not the measured investments of a maturing company. These are the desperate purchases of organizations terrified of being outpaced by competitors, particularly OpenAI. As Eli Mansoor, CEO of OskaQ Consulting, correctly notes: 'You will need to be in a rush to get infrastructure, and if you are not going to be there, you will be out of business.' That panic-driven reasoning should disqualify these commitments, not justify them.

The most alarming detail is Anthropic's decade-long contract with Amazon—double the typical five-year standard. As Corey Quinn, chief cloud economist at Duckbill, wisely observes, anything past three to four years means 'you're basically throwing darts in the dark' and 'it is almost a certainty that it will be renegotiated.' Anthropic has locked itself into prices and terms that may look absurd within years. Adding insult to injury, a data center lease starting in 2027 carries a 20-year term. What happens when AI needs change? When better technologies emerge? When compute costs plummet?

The fact that major tech companies—Advanced Micro Devices, Broadcom, Google, Microsoft, Nvidia, and Taiwan Semiconductor Manufacturing Company—are collectively investing $26 billion at the IPO price reveals the real dynamic here. They are not investing in Anthropic's success. They are investing in their own survival and revenue streams. Google's 13% stake and backstop of $19 billion in Terawulf lease commitments isn't confidence; it's insurance against a bet gone wrong.

Yes, investors will likely see short-term gains. But the tech industry is repeating mistakes it has made before: mistaking growth for profitability, confusing scale for superiority, and assuming exponential trajectories will continue indefinitely. The Nasdaq is down 5.3% in October alone, and political headwinds threaten data center development nationwide. Nvidia's leadership crisis should be another warning sign that even the sector's titans are vulnerable.

Anthropic needs to prove its Claude AI can justify its extraordinary spending. But going public now, with mounting market pressure and geopolitical uncertainty, suggests the company—and its backers—are more interested in liquidity than longevity. That's a recipe for value destruction, not creation.