AI Data-Center Revenue Leaves a Calculated Gap of at Least $4.9 Trillion

AI data-center owners would need at least $4.9 trillion more in cumulative revenue through 2030 than the projected ceiling for enterprise and consumer AI revenue tied to those facilities to cover announced spending, financing costs and a 12% return.

The comparison uses $8.5 trillion in cumulative revenue needed to cover announced capital spending and financing costs while delivering a 12% return on invested capital. It subtracts a forecast ceiling of $3.6 trillion in enterprise and consumer AI revenue attributed to data centers. Since that revenue figure is an upper limit, the resulting gap is at least $4.9 trillion.

The $3.6 trillion estimate covers revenue linked to data-center infrastructure, not revenue that facility owners would necessarily collect. Gartner Senior Director Analyst Will Sommer said “less than half of AI revenue will be earned by providers that own significant AI infrastructure.”

AI providers are expected to spend nearly $1 trillion on infrastructure in 2026, followed by another $6 trillion from 2027 through 2030. The $8.5 trillion requirement is cumulative through 2030 and includes both announced investment and financing costs.

If providers maintain their current spending pace, more than $3.5 trillion in AI infrastructure spending could be exposed to impairment by the end of 2030. Separately, more than $4 trillion in asset value could face write-down risk through 2032. These are projected risks, not losses already recorded.

The 12% return threshold is not expected to be reached before the end of 2032. The impairment estimate and the asset write-down estimate cover different periods and describe separate exposures.

More than 80% of neocloud providers, which supply AI-focused cloud infrastructure, are projected to stop operating independently by 2030. Mounting costs and a lack of distinct business offerings are among the pressures in that outlook. Spending on AI-optimized infrastructure-as-a-service is projected at $66 billion in 2027 and $174 billion in 2030.

A previous neocloud stock screen analysis examined the potential impairment risk.