Crusoe CEO Chase Lochmiller: Turning Stranded Energy Into Intelligence
Chase Lochmiller, founder and CEO of Crusoe, discusses building a vertically integrated AI infrastructure company spanning energy, data centers, manufacturing, and managed AI cloud. Topics include leading a complex organization, the Bitcoin-to-AI pivot, energy-first compute, executive coaching, firing decisions, community pushback on data centers, and lessons from climbing Everest.
Building a Vertically Integrated AI Infrastructure Company
Crusoe is not a cloud provider in the conventional sense. Chase Lochmiller describes a business that spans energy generation, power plant construction, electrical equipment manufacturing, gigawatt-scale data centers, and a managed AI cloud with managed AI services. The company’s model is closer to an oil major than a software firm: upstream energy, midstream infrastructure, and downstream compute, all integrated so that margins in one layer can offset pressure in another.
Scaling Leadership Across Domains
Lochmiller’s answer to the “dizzyingly complex” problem is to rely on domain experts in each vertical (energy, data centers, manufacturing, and AI cloud) while staying close enough to each one to go deep when needed. He describes the arrangement as trust with accountability: leaders are aligned on direction, understand how their piece affects the rest of the business, and are then held responsible for executing the agreed plan. He notes that the boiling frog effect helps founders, since they carried the company’s complexity up gradually from a two-person start, while newcomers arrive to a company that already feels intense.
He also points to the limits of this approach. Directly responsible individuals (DRIs) for cross-organization work are hard to find, and his own estimate is that most leaders have trust for only three or four people across the whole organization. Cross-functional delivery is a persistent struggle. His main tool is a weekly staff meeting where pain points between teams surface and get addressed. His more striking observation is about repetition: as the organization grows, he finds himself saying the same things until he is sick of hearing them, and that repetition is what alignment costs at scale.
Urgency and the Fight Against Inertia
Lochmiller argues that large companies lose the speed that makes startups effective, and that resisting this inertia is a deliberate job. He admires Elon Musk’s ability to pair a vision of changing the world with a sense that the company could fail tomorrow, and says he himself applies that tension less extremely. On competition, he cites advice that startups die of suicide, not homicide: they fail through poor execution, weak culture, and the wrong hires, not because a rival wins. He treats competitors as validation of a real market and sees infrastructure as a layer with room for many winners.
Energy First
The founding thesis was energy first. Crusoe began by finding stranded resources, most notably natural gas flared at oil fields, and bringing compute to that energy rather than moving energy to a market. Bitcoin mining was the first monetization engine because it was the simplest way to turn energy into revenue. The company has since worked with behind-the-meter wind, gas, solar, and batteries, geothermal and hydro in Iceland and Norway, a partnership on an SMR-powered AI factory at Idaho National Lab, and Blue Energy’s nuclear work. Lochmiller frames the energy mix as a spectrum, not a silver bullet: he expects different solutions in different places, and sees battery innovation (including second-life EV batteries and long-duration storage) as part of the answer.
From Bitcoin to AI
Lochmiller says the AI pivot was the plan from the start, not a reaction to a market shift. His reasoning draws on early conviction that AI is a general-purpose technology, like electricity, and that if data and compute are the bottlenecks, compute becomes an energy problem. Bitcoin and AI pair well because AI load is more variable, with training spikes and daytime cooling demand, while mining can run at a flatter draw. The turning point was the launch of ChatGPT in November 2022, which created demand for the infrastructure Crusoe had already been building. The decision to divest the Bitcoin business, sold to a team with deep natural gas experience, came down to focus: capital and attention were being split, and investors were restricted from crypto-linked investments, which complicated the story. He remains a Bitcoin holder and supporter.
Leadership Development and Hiring
Lochmiller describes public speaking as a skill he had to build deliberately, having come from a secretive quant background. A coach was central to that growth, and he describes the relationship as part coach and part therapist, built on trust that compounds over time. He praises Felicis for a self-improvement budget funded for founders, reasoning that venture bets rest on founders’ ability to grow. On people decisions, he argues that most leaders fire too slowly and that every case is different. He cites the observation that people scale linearly while companies scale exponentially, so the person who fit a role at 200 people may not fit at 2,000. He acknowledges the tension between caring personally about people and making the right business call, and frames the choice as a judgment that combines many small signals from across the organization.
Culture Across Welders and Researchers
Crusoe’s culture rests on a value of tapping into collective genius. Lochmiller argues that no single person holds all the expertise needed, and gives examples of electricians whose field knowledge improved the work of AI researchers. He sees the full stack, from the electron to the token, as a source of differentiation.
Public Criticism and Dealing With Bad News
On community opposition to data centers, Lochmiller’s position is that the industry’s biggest problem is marketing: the truth has not been communicated well. He describes meeting county officials and residents directly, leading with facts rather than superiority, and finds that people respond to authenticity independent of political allegiance. He also argues that thick skin is essential for a CEO, since the job is largely receiving bad news. His approach is to actively mine for criticism, reward candor, and stay approachable as the company grows.
Everest and Conscious Risk
His 2014 Everest attempt ended after an avalanche in the Khumbu Icefall killed 15 Sherpas. Returning in 2018, he and his partner redesigned the expedition to minimize the exposure they could not control: they acclimatized on nearby peaks, making one major acclimatization trip up Everest instead of the usual sequence, cutting their trips through the icefall from eight passes to four. The principle he draws is the same one he applies to business, which is to accept risks consciously and reduce the stochastic ones. The summit itself came after a difficult night near the Balcony, where a partner’s encouragement got him through, and sunrise brought a sense of awe and perspective.