← Latest briefing

Politics

Campaigns spend millions targeting data center concerns ahead of midterm elections

Both parties are using public frustration over rising utility bills to target artificial intelligence facilities in competitive races.

The short version

  • Campaigns and outside groups have spent more than $45 million on advertisements mentioning data centers across gubernatorial, House, and Senate races.[NPR]
  • Spending is almost evenly split between parties, with over $22 million backing Republicans and over $21 million backing Democrats.[NPR]
  • A June poll showed 77% of Americans are concerned that data centers raise electricity costs, spurring bipartisan focus in local and state contests.[NPR]
  • Gubernatorial races lead total spending at about $32 million, led by Georgia, while an internal Senate memo warned a tech loss could heighten nationwide pushback.[NPR]

Key facts

  • Political campaigns and political groups have spent over $45 million on ads focusing on data centers in congressional and gubernatorial races since January.[NPR]
  • Ad spending was nearly identical across party lines through August, with more than $22 million supporting Republican candidates and over $21 million supporting Democrats.[NPR]
  • Gubernatorial contests led all ad expenditures on data centers at about $32 million, followed by $8 million in House races and $5.3 million in Senate races.[NPR]
  • A June Reuters/Ipsos survey indicated that 77% of Americans worry about data centers increasing electricity rates, with opposition found across both party bases.[NPR]
  • Georgia accounted for roughly 30% of total data center ad spending and over half of Republican expenditures on the topic.[NPR]

What remains uncertain

  • A Republican Party operative noted that the party lacks a unified public stance because internal views on data centers vary between economic competitiveness and ratepayer protections.[NPR]

Sources

Outlet counts describe coverage, not independent confirmation. Reports may share a wire service or original source.