Data Center Efficiency and Building Decisions

AI is destroying the planet—or so we've been told. This week on Chain of Thought, we tackle one of the most persistent and misleading narratives in the AI conversation.

Andy Masley, Director of Effective Altruism DC, joins host Conor Bronsdon to fact-check the absurd AI environmental claims you've heard at parties, in articles, and even in bestselling books. Andy recently went viral for discovering what he calls "the single most egregious math mistake" he's ever seen in a book—a data center water usage calculation in Karen Hao's NYT Bestseller, Empire of AI, that was off by a factor of 4,500.

In this conversation, Andy and Conor break down the myths around AI’s water and energy usage and explore:

• The viral Empire of AI error and what it reveals about the broader debate

• Why most AI water usage statistics are misleading or flat-out wrong

• How one ChatGPT prompt represents just 1/150,000th of your daily emissions

• Trade-offs around data center cooling + decision making

• Why "tribal thinking" about AI is distorting environmental activism

• Where AI might actually help the climate through deep learning optimization

If you've ever felt guilty about using AI tools, been cornered at a party about AI's environmental impact, or simply want to understand what the data actually says, this episode, and Andy’s deep dive articles, arm you with the facts.

Chapters:

00:00 – Introduction: The Party Guilt Problem

01:54 – Andy's Background and What Sparked This Work

03:50 – The 4,500x Error in Empire of AI

06:39 – Breaking Down the Math: Liters vs. Cubic Meters

10:39 – The Unintended Consequence: Air Cooling vs. Water Cooling

12:51 – Karen Hao's Response and What's Still Missing

19:08 – Why Environmentalists Should Focus Elsewhere

21:41 – The Danger of Tribal Thinking About AI

25:49 – What Is Effective Altruism (And Why People Attack It)

29:15 – EA, AI Risk, and P(doom)

34:31 – Why Misinformation Hurts Your Own Side

37:39 – Using ChatGPT Is Not Bad for the Environment

42:14 – The Party Rebuttal: Practical Comparisons

45:23 – Water Use Reality: 1/800,000th of Your Daily Footprint

48:27 – The Personal Carbon Footprint Distraction

53:38 – Data Centers: Efficiency vs. Whether to Build

55:13 – AI's Net Climate Impact: The Positive Case

59:34 – Deep Learning, Smart Grids, and Climate Optimization

1:03:45 – Final Thoughts

Key references

IEA Study: AI and climate change - https://www.iea.org/reports/energy-and-ai/ai-and-climate-change#abstract 

Nature: https://www.nature.com/articles/s44168-025-00252-3 

The Empire of AI Error: https://andymasley.substack.com/p/empire-of-ai-is-wildly-misleading 

Using ChatGPT isn’t bad for the environment: https://andymasley.substack.com/p/a-short-summary-of-my-argument-that

https://andymasley.substack.com/p/a-cheat-sheet-for-conversations-about 

Connect with Andy Masley: 

Substack – https://andymasley.substack.com/

X (Twitter) – https://x.com/AndyMasley

Connect with Conor Bronsdon: 

Substack – https://conorbronsdon.substack.com/

LinkedIn – https://www.linkedin.com/in/conorbronsdon/

X (Twitter) – https://x.com/ConorBronsdon