Anthropic IPO leads AI's $235 billion dealmaking blitz

SpaceX recently secured $40 billion in debt financing specifically for leasing GPUs, the essential chips powering modern artificial intelligence. Concurrently, Meta indicated plans to seek up to $95 billion in funding for its own AI development efforts.

These transactions, alongside Anthropic's recent filing for what might be the largest initial public offering ever, represent a staggering $235 billion mobilized in a single day. This figure surpasses the entire economic contribution of the United States airline industry within one year.

The intense demand for capital in this sector has clearly surpassed the capacity of any single segment of the financial world to supply. Anthropic alone projects commitments of around $750 billion toward computing power over the coming years, a sum seven times greater than its potential IPO raise.

Anthropic submitted its paperwork yesterday, potentially aiming for a record $100 billion. This filing provides a detailed look at the firm’s financials and associated perils as it prepares for its market debut.

The prospectus details that the firm has nearly 2,000 clients generating $1 million or more in revenue, with 250 of those clients spending $10 million or more. One analyst noted, "We are so early," referencing the $1.5 trillion annual enterprise software expenditure by global businesses outside China.

Despite this expansion, the company reported a net loss of $80.7 billion in the first half of 2026, a steep increase from the $13 billion loss seen during the same period the prior year.

The documents also carry severe warnings, stating that AI might present "catastrophic or existential risks to humanity," and that its sophisticated models could attempt to manipulate users.

Furthermore, the paperwork reveals that Anthropic's payments to SpaceX are climbing, now projected to reach $3.1 billion monthly, potentially totaling up to $120.5 billion in computing through 2029.