How the AI Oligarchy Went Hyperscale (with Tim Murphy)

How the AI Oligarchy Went Hyperscale (with Tim Murphy)

From Pitchfork Economics with Nick Hanauer by Civic Ventures

May 12, 2026 · 39 min

About this episode

Nick Hanauer and Tim Murphy discuss the impact of hyperscale data centers on communities and the economic model behind the AI boom.

The AI “cloud” sounds weightless. But behind every chat bot, every prompt, and every promise of a coming AI revolution is a massive physical footprint: hyperscale data centers consuming enormous amounts of land, electricity, water, and public subsidies. This week, Nick and Goldy talk with Tim Murphy, national correspondent at Mother Jones, about his cover story on how the American oligarchy went hyperscale in the age of AI. Murphy has been reporting from communities across the country where residents are watching enormous data centers rise in their backyards, often with little transparency, few long-term jobs, and huge demands on local infrastructure. The result is a familiar story: public risk, private reward. Tech billionaires get the profits. Communities get higher utility costs, depleted resources, tax breaks they may never recoup, and facilities that could become tomorrow’s stranded assets when the AI bubble bursts. AI may be new. But the economic model behind this boom is very old: extract from communities, concentrate power at the top, and call it progress. Tim Murphy is a national correspondent at Mother Jones. Social Media: @timothypmurphy.bsky.social @timothypmurphy…

People in this episode

Host: Nick Hanauer

Guest: Tim Murphy

Topics covered

  • AI
  • oligarchy
  • data centers
  • public infrastructure
  • economic model

Keywords

  • AI revolution
  • data centers
  • public risk
  • private reward
  • economic model

Mentioned in this episode

Organizations: Mother Jones, Civic Ventures

More episodes of Pitchfork Economics with Nick Hanauer

Explore listener stats, chart rankings, contacts and more on the Pitchfork Economics with Nick Hanauer podcast page.