Power Grid Constraints Drive Next AI Trade Boom

Introduction

The AI boom has been framed as a race for chips, but UBS thinks the bottleneck has moved to the power grid.

The Power Grid Constraint

UBS argues that infrastructure constraints, specifically power supply, grid capacity, and data center space, are driving the next phase of the AI trade.

The Timeline Mismatch

UBS notes that a data center can go up in 12 to 24 months, but adding power capacity can take 5 to 10 years.

The Bill for Closing the Gap

UBS estimates $511 billion will be needed for new US power-generation capacity by 2030 to support a 3% compound annual growth rate in electricity demand.

Equipment and Policy as Choke Points

Large transformers may face lead times of up to 210 weeks, and policy is adding friction too, with some states requiring developers to pay for grid upgrades themselves.

Opportunities for Growth

Despite the challenges, UBS sees opportunity in the colocation data center stocks, with around 80% annual EBITDA growth projected from 2027 to 2029.

Conclusion

A pipeline of 276 GW of planned data center capacity looks impressive, but the practical watch list follows from the report, focusing on regional regulatory moves, interconnection queue trends, equipment lead times, and power project timelines.