Catalyst with Shayle Kann
Catalyst with Shayle Kann

Live from Transition-AI 2026: Inside Google’s massive AI CapEx

April 23, 2026

AI Summary

5 min read

Google’s AI infrastructure chief says the biggest constraint on AI growth isn’t just chips or power or labor—it’s all of them, simultaneously, every single day. “At 10 a.m. it’s labor, at noon it’s power, and at 2 p.m. it’s chips,” says Amin Vadat, Google’s Chief Technologist for AI infrastructure, speaking live at the Transition-AI 2026 conference. His comment captures the multidimensional bottleneck facing the industry as Google prepares to spend $175–185 billion in CapEx this year—roughly seven times NASA’s annual budget or five times the cost of the Vogtle nuclear plant.

The inference transition changes data center scale

The era of ever-larger training clusters is giving way to something more nuanced. Vadat explains that Google’s first data center in Oregon was a then-astonishing 10 megawatts. Today, the company builds gigawatt-scale facilities. But as AI shifts from training to inference—serving models to users—the need for individual data center scale changes fundamentally.

Continue reading the full summary in the app — free to try.

Read Full Summary →

Free • No credit card required

Listen to Audio Summary Open in App

What you'll learn

  • 1 (01:49) **Context: Google’s AI CapEx and the Scale Question** - Shayle frames the conversation with Google’s $175-185B CapEx announcement and asks how much individual data center scale matters for inference vs. training.
  • 2 (10:23) **Site Economics: Large vs. Medium Data Centers** - Shayle asks whether building one gigawatt site or many smaller sites will be easier in the future.
  • 3 (11:52) **Why Data Centers Demand High Reliability—and Why That’s Changing** - Shayle asks why data centers require 4 nines reliability and if that’s intrinsic.
  • 4 (14:43) **Behind-the-Meter Power: Bridge or Permanent?** - Shayle asks about the trend toward on-site generation and storage.
  • 5 (17:07) **The Bridge Power Dilemma: Over-Provisioning On-Site** - Shayle probes the reliability challenge of using on-site generation during the bridge period.
  • 6 (18:10) **Stranded On-Site Generation?** - Shayle asks if excess on-site capacity will become stranded once grid connections arrive.
  • 7 (21:17) **Microgrids and Software Control** - Shayle asks if data centers with on-site generation will look like microgrids, co-optimizing multiple resources.

+ Full timestamped outline available in the app

Show Notes

As the race to build out artificial intelligence accelerates, the infrastructure required to support it is undergoing a remarkable transformation. In February, Google announced a plan to spend $175 billion to $185 billion in CapEx for 2026— a figure roughly equivalent to the GDP of Hungary.

In this special live episode, recorded at Transition-AI 2026 in San Francisco, Shayle sits down with Amin Vahdat, Google’s chief technologist for AI infrastructure. Amin pulls back the curtain on how the hyperscaler is rethinking everything from data center reliability and behind-the-meter power generation to real-time inference.

Shayle and Amin discuss:

  • How Google’s shift from focusing on training to inference can enable more distributed, smaller-scale data center deployments
  • Why Google is moving away from traditional "five nines" reliability for certain workloads in exchange for doubling compute capacity
  • How on-site generation can serve as a "bridge" to manage interconnection latency
  • Google’s milestone agreement with utilities for one gigawatt of demand response
  • How software can co-optimize chip design, building cooling and power generation to create superefficient and flexible "AI factories"
  • Catalyst: The rise of flexible data centers
  • Catalyst: Will inference move to the edge?
  • Catalyst: The mechanics of data center flexibility
  • Open Circuit: The natural gas ‘bridge’ becomes a highway
  • Open Circuit: Are investors losing faith in the AI infrastructure frenzy?
  • Latitude Media: Energy Vault is expanding into infrastructure for AI
  • Latitude Media: The rise of the AI infrastructure asset class


Credits: Hosted by Shayle Kann. Produced and edited by Max Savage Levenson. Original music and engineering by Sean Marquand. Stephen Lacey is our executive editor.

Catalyst is brought to you by FischTank PR, an award-winning climate and energy tech, renewables, and sustainability-focused PR firm dedicated to elevating the work of both early-stage and established companies. Learn more about their PR approach

Catalyst with Shayle Kann

More from this podcast

Catalyst with Shayle Kann →