Alphabet Pays $4.75B to Solve AI's Power Problem
Google's parent company acquires data center and energy infrastructure firm Intersect to secure electricity for its AI ambitions as U.S. grids struggle to keep pace.
Google is buying its way out of a power crunch.
Alphabet announced a definitive agreement to acquire Intersect, a data center and energy infrastructure company, for $4.75 billion in cash plus assumed debt. The deal is expected to close in the first half of 2026.
The acquisition addresses a fundamental constraint facing every major AI player: electricity. U.S. power grids are struggling to meet demand that's surging for the first time in decades, driven largely by the computational requirements of generative AI.
The Strategic Logic
Intersect has $15 billion of assets in operation or under construction across the U.S. The company specializes in co-locating data centers with dedicated power generation—a model that bypasses grid bottlenecks entirely.
Alphabet already held a minority stake from an $800 million funding round last December. That partnership set a target of $20 billion in total investment by 2030.
Sundar Pichai put it plainly: "Intersect will help us expand capacity, operate more nimbly in building new power generation in lockstep with new data center load, and reimagine energy solutions to drive US innovation and leadership."
What Stays Independent
The deal includes important carve-outs. Intersect's existing operating assets in Texas and its California portfolio will remain with an independent company backed by TPG Rise Climate and other investors.
Intersect's operations will stay separate from Google under the Intersect brand, led by CEO Sheldon Kimber. The first joint project—a co-located data center and power site in Haskell County, Texas—is already under construction.
The AI Infrastructure Race
Every hyperscaler is scrambling for power. Microsoft has deals with nuclear plants. Amazon is buying data center campuses with dedicated energy sources. Google's approach—acquiring an integrated infrastructure developer—offers more control over the timeline.
The $4.75 billion price tag is substantial but not surprising given what's at stake. For Alphabet, the alternative is waiting in line for grid capacity that may take years to materialize.
AI leadership increasingly depends on megawatts, not just algorithms.