Big Tech Can't Stop AI Spending, but the Cash Is Running Out

Big Tech's escalating investment in artificial intelligence infrastructure is creating significant financial strain, with combined capital spending for Microsoft, Alphabet, Amazon, Meta, and Oracle projected to hit $945.55 billion by January next year. This represents a six-fold increase from January 2023, while their free cash flow is expected to swing into a $2.87 billion deficit from a $223.11 billion surplus over the same period. The market squeeze has reportedly exhausted traditional funding avenues like internal cash and corporate bond issuance.

Meta Platforms' AI infrastructure commitment, totaling nearly $825 billion, has particularly chilled the market. This figure, which includes $355.7 billion in off-balance-sheet operating and finance leases and $454.5 billion in non-cancelable contracts, increased by $125 billion in just three months. Immediately following this disclosure, the premium on Meta's credit default swaps, an indicator of default risk, jumped to 119.3 basis points, signaling heightened market anxiety.

In response to mounting funding pressure, major players are turning to new financing strategies. Meta plans to issue non-voting Class C common shares in January, distributing new shares to existing holders and then selling additional shares into the market to raise capital. This approach allows the company to secure funds without diluting the voting power of current shareholders.

Similarly, both Anthropic and OpenAI are pressing ahead with initial public offerings. Anthropic, which spent $9 billion on computing capacity in the first half of this year and faces up to $750 billion in future contractual obligations, plans its IPO next month. OpenAI is also proceeding with its IPO next year, even after reportedly lowering its valuation target to $1.25 trillion due to investor feedback, indicating a pragmatic adjustment to market conditions in its pursuit of capital.