Who Will Pay the $2.6 Trillion AI Infrastructure Bill?

BCG's cost-model analysis estimates that AI infrastructure will need roughly $2.6 trillion in annual spending by 2030 and maps how enterprises and consumers can ultimately pay for that buildout, plus which signals to monitor. The report frames who bears the cost of frontier-scale AI infrastructure — the financing gap and the burden-sharing between suppliers, enterprises and consumers — rather than treating capex as an exogenous given.

For CFOs and strategy teams, this quantifies the scale and financing structure of the AI infrastructure build-out, informing capex-planning assumptions and risk of a build-out pause.

Hand to the CFO/strategy team to pressure-test AI infrastructure investment assumptions against BCG's $2.6T-by-2030 cost model.