Google, Microsoft, Meta, and Amazon collectively committed approximately $700 billion to AI infrastructure spending this year, a 77% increase over 2025. All four reported earnings this week. Every single one beat expectations. The stock market yawned.
Google Cloud crossed $20 billion in quarterly revenue and told analysts it would have grown faster if it had more capacity. An $80 billion annual run-rate business said it left money on the table because it couldn’t build data centers fast enough. The money is going to chips, data centers, and power. Microsoft pledged $1 billion to restart Three Mile Island’s Unit 2 nuclear reactor. Meta has RFPs out for several small modular nuclear plants. Nobody restarts a nuclear reactor for a quarterly earnings bump.
This feels like a classic game of prisoner’s dilemma. If Google stops building and Microsoft doesn’t, Google loses enterprise AI workloads permanently. If Meta stops and Google doesn’t, Meta can’t serve AI features to three billion users. The rational move for each player is to keep spending, regardless of what the collective total looks like.
Big Tech is calling this right: the companies that kept laying railroad tracks in the 1870s owned the next century, and the ones that flinched became footnotes. Can you afford to be the one who stops? The bears call $700 billion irrational. The demand curve says otherwise.
Every company needs a Claw strategy. Do you have one?
Author’s note: This is not a sponsored post. I am the author of this article and it expresses my own opinions. I am not, nor is my company, receiving compensation for it. This work was created with the assistance of various generative AI models.