Thanksgiving Week RepostThis episode originally aired in June 2024. We’re resurfacing it because the core idea discussed here were timely then and even more timely now.

We’ve also refreshed the audio, with improved mixing and mastering for a clearer, smoother listen.

Crusoe has scaled dramatically since this conversation, including major new funding and new projects in Texas. With so much energy news focused on problems, it felt right this week to highlight solutions in action.

When most people see flares in the Permian, they wonder why all that energy is being wasted. Crusoe’s co-founders figured out how to put that wasted energy to good use. They started with cryptocurrency mining and have steadily moved to AI data centers.

Over the last few years, they have found themselves perfectly positioned to grow as the AI boom took hold. They’ve recently completed the 8th building at Stargate in Abilene for Open AI and Oracle. They’re also building facilities for Google near Amarillo.

In this conversation from May 2024, Crusoe Co-Founder, President, and COO Cully Cavness and I talked about the rapidly growing size of data centers, the flexibility of different kinds of data centers, and how large loads can increase grid reliability.

This was one of the earlier podcasts on these topics and I think it holds up really well. For those looking for more on the topic, here are some other Energy Capital Podcasts covering similar ground:

What Has Changed Since Then

When this episode first aired in mid-2024, Crusoe was already shifting from “flare mitigation plus computing” to a broader energy-first AI infrastructure model.

In the time since:

* Crusoe has become one of the most aggressive builders of AI data centers in the country. It is now described as an “AI factory company” with a vertically integrated cloud platform built around stranded and low-cost energy.

* Abilene, Texas moved from concept to centerpiece. Crusoe is building a 1.2 gigawatt data center at the Lancium Clean Campus outside Abilene — Stargate — as the first phase of a planned 5 GW campus.

* The company’s capital and pipeline exploded. Since 2024, Crusoe has raised hundreds of millions of dollars to scale “clean energy data centers,” then a further $1.3 billion in Series E financing, bringing total funding close to $4 billion and valuing the company around $10 billion.

In other words, the approach Cully describes in this episode has scaled — rapidly.

Why The Core Idea Still Matters For Texas

The heart of the episode is simple:

* Methane mitigation is still some of the lowest-hanging fruit in climate policy. Crusoe’s digital flare mitigation aims for 99 percent plus combustion efficiency, cutting the climate impact of flaring while turning waste into power.

* Curtailment and congestion are still big problems in West Texas. A “go to the energy” model lets data centers soak up low-priced or stranded wind and solar instead of forcing renewable operators to shut down when prices go negative.

* AI loads can be designed to help rather than hurt the grid. Some training workloads can be paused or shifted toward hours when renewables are plentiful. That kind of flexibility is exactly what ERCOT needs as large loads and renewables grow together.

Texas sits at the center of all three issues. We flare and vent more than we should. We waste clean power when transmission is full. We are a magnet for AI and industrial loads.

Crusoe’s solutions help with all of these challenges.

Final Thoughts

This episode is worth revisiting because it offers a concrete picture of one possible future for Texas: fewer wasted molecules, less wasted renewable power, and more large loads designed with the grid in mind.

If you listen again with today’s headlines in mind, I would be interested to hear what stands out for you. If you know someone working in oil and gas, renewables, or AI infrastructure in Texas, feel free to share it with them.

We will not get every siting decision right. But we do have choices about whether AI growth deepens our problems or helps solve them.

Timestamps

* 00:00 – Introduction

* 01:30 – Cully’s background and the origin story of Crusoe

* 08:00 – How digital flare mitigation works and why it cuts methane emissions

* 15:00 – Digital renewables optimization, negative pricing, & stranded wind power

* 21:00 – Data center and AI demand growth and what it means for the grid

* 28:00– Flexibility of AI workloads and how data centers can act as flexible loads

* 38:00 – Efficiency gains in AI chips and power density in modern racks

* 41:00 – Location-based versus market-based carbon accounting

* 43:00 – “Tally’s Law” and what it tells us about the energy transition

* 50:00 – Policy and regulatory changes that could accelerate this kind of solution

Show Notes

Host, Guest, & Company• Cully Cavness - LinkedIn, Twitter/X• Crusoe Energy - Crusoe Careers Page - LinkedIn, Twitter/X• Doug Lewin - LinkedIn, Twitter(X), Bluesky, & YouTube

Mentions in the Podcast:• Tally’s Law and the Energy Transition by Cully Cavness• The Extraction State by Charles Blanchard (book)• AI, Data Centers & Energy, Interview w/ Michael Terrell - Redefining Energy Podcast• AI is poised to drive 160% increase in data center power demand - Report from Goldman Sachs• Nuclear? Perhaps! - Interview with Jigar Shah on the Volts Podcast• Texas Advanced Nuclear Reactor Working Group at the Texas Public Utility Commission

Transcript

Doug Lewin

Cully Cavness, welcome to the Energy Capital Podcast.

Cully Cavness

Thank you so much for having me.

Doug Lewin

Really looking forward to this conversation. Crusoe is really a fascinating company. You guys are doing some really innovative, interesting, and different things. So why don’t we start with you, Cully? Tell us a little bit about your background and about Crusoe. Explain to the audience a little bit who you guys are as a company, if you would.

Cully Cavness

Great. I’m excited to be here and share a little bit about what we’re doing at Crusoe, where we came from, where we’re going. In terms of my personal background, I grew up in Denver, Colorado. I went to Middlebury College in Vermont to study and I studied geology and economics thinking I was gonna go into oil and gas. But at Middlebury, anybody who’s familiar with the school will know that the climate conversation was a huge theme and a huge focus in that student body. And it made a big impact on me.

And so I actually, right after I graduated from college, I was awarded a Thomas Watson Fellowship, which is a program where you’re sort of banished from your home country for a year and you get to go study whatever subject you really want to study for that year. And I wanted to think about this sort of morality of energy and the balance between energy and the economy and the environment. And so I was really fortunate to be able to go to Iceland where I worked with a lot of geothermal power and hydro producers. I went to China where I was much closer to coal. And then I went to Spain. I worked with wind and solar developers for the CFO of a large renewables group there. And then I went to Argentina and I worked wi...

Replay: Using Wasted Energy to Power AI with Crusoe's Cully Cavness