Gartner warns AI infrastructure spending will fall $4.9 trillion short of needed revenue

Gartner’s latest press release highlights a widening gap between AI infrastructure spending and the revenue it can generate. Data-center owners need to earn $8.5 trillion in cumulative revenue by the end of the decade to meet a 12% return hurdle, yet Gartner estimates only $3.6 trillion of enterprise and consumer AI revenue will be attributable to those centers, creating a $4.9 trillion deficit. Annual AI capital spending is projected to approach $2 trillion by 2030, while total AI software, services and consumer product revenue is expected to hit $4.4 trillion, with merely 30% tied to underlying infrastructure.

The firm cautions that over $3.5 trillion of AI infrastructure could be written down by 2030, and more than $4 trillion by 2032. Large cloud operators like Amazon, Microsoft and Alphabet have diversified revenue streams to absorb the pressure, but smaller “neocloud” providers may see over 80% exit the market by 2030 due to cost pressures and lack of differentiation. Gartner advises product leaders to assume a more concentrated AI supply landscape and to factor potential write-downs into product economics.