Why the Father of Fracking Says Oil and Gas Are the Only Fuels That Can Keep AI Running
If you’re in B2B SaaS, you’ve heard the AI gold rush stories. But there’s one bottleneck no one in San Francisco wants to talk about: the grid.
I just sat in the audience at America Innovates in San Francisco, California, where Kerry Dolan interviewed a man who literally rewired the global energy map. Harold Hamm—the pioneer of hydraulic fracturing—didn’t come to talk about drilling. He came to talk about the biggest threat to AI’s future.
And his message was blunt: Oil and natural gas are the only reliable energy sources capable of powering the AI revolution.
This isn’t a political take. It’s a math problem. And if you’re building a GTM engine, hiring a sales team, or raising a Series B for an AI-native product, you need to understand what Hamm said. Because energy infrastructure will dictate where data centers get built, how much compute costs, and ultimately, which SaaS platforms can scale.
Let’s break down Hamm’s core argument—and what it means for revenue teams in 2026.
The AI Energy Paradox: More Compute, Less Grid
Here’s the reality that most tech founders don’t internalize until they try to buy GPU capacity:
- A single ChatGPT query consumes roughly 10x the energy of a standard Google search.
- Training a frontier model like GPT-5 (or its successors) requires megawatt-hours of continuous power.
- Hyperscalers—AWS, Azure, Google Cloud—are racing to secure gigawatts of capacity. But the U.S. grid wasn’t built for this.
Harold Hamm made it simple during the America Innovates interview: “You can’t run AI on intermittent energy.”
Translation: Solar and wind are amazing. They’re also unreliable when a cloud rolls over or the wind dies. Data centers need 24/7/365 baseload power. And right now, only two sources can deliver that at scale—natural gas and nuclear.
Nuclear is coming, but it takes a decade to permit a plant. We don’t have a decade.
That leaves natural gas. And Hamm is betting his legacy on it.
Why Harold Hamm’s Voice Matters Right Now
If you’re under 40, you might not know the name. But here’s the context: Harold Hamm is the father of the shale revolution. In the 2000s, he pioneered horizontal drilling and hydraulic fracturing techniques that unlocked trillions of cubic feet of natural gas from the Bakken formation in North Dakota.
Before Hamm, the U.S. was an energy importer. After Hamm, the U.S. became the world’s largest producer of oil and gas. That’s not hype. That’s economic history.
So when Hamm says oil and gas are the only reliable fuels for AI, he’s not guessing. He’s done the math on deliverability.
At America Innovates, he laid out the core tension:
- Renewables are great for peaking power.
- Natural gas is the backbone for baseload.
And here’s the kicker: Natural gas-fired power plants can be built in 2–3 years. Permitting is faster than nuclear. The infrastructure (pipelines, processing plants) already exists in most of the country.
That speed is everything when you’re trying to deploy a data center before your competitor does.
The GTM Playbook: What Revenue Teams Should Watch
If you sell B2B SaaS to any of the following segments, Harold Hamm’s energy thesis is a signal you cannot ignore.
1. Data Center Infrastructure & Colocation
The biggest cloud providers (AWS, Microsoft, Google) are already signing long-term power purchase agreements (PPAs) with natural gas producers. Expect more.
What this means for your sales motion:
- Data center sales cycles are compressing because power availability is the primary constraint. If your tool speeds up site selection, energy procurement, or construction permitting, you have a wedge.
- Pricing power is shifting. Data center operators who secure cheap, reliable gas will have lower cost bases. They’ll pass that to customers (you) or keep margin.
Actionable tip: Go listen to earnings calls for Equinix, Digital Realty, or CyrusOne. Every single one now mentions “power availability” as a top risk factor. That’s your opening.
2. AI-Native SaaS Companies
If you’re building a platform that relies on large language models or real-time inference, your cloud bill is about to be a recurring nightmare.
Why Hamm’s thesis matters:
- In regions where gas is abundant (Texas, Louisiana, Pennsylvania), data center power costs are 20–30% lower than in California or the Northeast.
- If your startup hasn’t already mapped AWS regions to local energy prices, you’re leaving margin on the table.
Revenue play: Offer compute cost optimization as a value prop. Your pricing model needs to flex based on where the models run. The companies that figure this out first will undercut competitors on price.
3. Energy Tech & CleanTech SaaS
This is the big one. Harold Hamm’s interview signals a pivot in messaging.
For years, clean tech startups pitched themselves as “renewable-first.” Hamm’s argument flips the script: natural gas is the bridge fuel that enables renewables to work.
What you should do:
- If your product helps O&G companies optimize drilling, reduce flaring, or monitor methane leaks, rebrand your messaging around AI enablement. You’re not an energy company—you’re an AI infrastructure company.
- Expect more VC dollars to flow into “dual-use” technologies that work with both gas and renewables. The winners will be platforms that help operators manage both.
- Start following the Non-Traditional Resource (NTR) trend. Hamm’s company, Continental Resources, is pioneering EOR (enhanced oil recovery) techniques that can extract more oil with less surface footprint. That’s a software opportunity.
The Data Behind the Dogma
Let’s ground this in numbers so you can use them in your next board meeting or pitch deck.
| Energy Source | Baseload Reliability | Build Time | Cost per MWh (LCOE) |
|---|---|---|---|
| Natural Gas | 95%+ | 2–3 years | $30–$50 |
| Solar | 20–25% | 1–2 years | $25–$40 (intermittent) |
| Wind | 30–40% | 2–3 years | $30–$60 (intermittent) |
| Nuclear | 90%+ | 7–10 years | $100–$150 |
| Coal | 85%+ | 4–6 years | $60–$80 (declining) |
Source: Lazard Levelized Cost of Energy Analysis + EIA data.
Notice the gap? Natural gas is the only source that combines high reliability, fast build times, and low cost. That’s Hamm’s whole argument.
For a 1GW data center campus (enough to power ~800,000 homes), you need ~4,000 MW of solar or wind to get the same reliable output as 1,000 MW of gas. And you need massive battery storage. That math doesn’t work today.
What Harold Hamm Got Right (and What His Critics Miss)
I’m not an energy apologist. I’ve seen the environmental impact of flaring and methane leaks. But Hamm’s point is pragmatic: We cannot build AI infrastructure on wishful thinking.
Critics will say:
- “Natural gas is still a fossil fuel.”
- “Carbon capture is unproven at scale.”
- “Battery storage will solve intermittency.”
Hamm’s response (paraphrased from the America Innovates stage):
“We’re going to need every molecule of energy we can produce. Oil and gas are the most reliable. We should use them to power the AI revolution while we build the next generation of clean energy. The two aren’t mutually exclusive.”
That’s not a contrarian take. That’s a realist take from someone who’s spent 50 years delivering energy at scale.
And for B2B leaders, realism is what you need when you’re planning a 5-year GTM strategy. Because nothing kills a sales motion faster than a power outage in your cloud region.
Three Takeaways for Your SaaS Strategy
1. Map your cloud regions to local energy mixes.
If your customers run workloads in AWS us-east-1 (Virginia), gas is your primary energy source. If they’re in eu-central-1 (Frankfurt), nuclear and gas dominate. Optimize your pricing and latency claims accordingly.
2. Start listening to energy policy.
The Inflation Reduction Act and infrastructure bills are driving billions into transmission lines. But Hamm’s argument suggests we’ll also see more federal support for natural gas pipelines. If you sell to utilities, this is your new wedge.
3. Build for reliability, not just speed.
Your customers care about uptime. The ones running AI inference on real-time data can’t afford a cloud outage because a solar farm went offline at night. Sell them on your reliability—backed by the same baseload logic Hamm is selling.
The Bottom Line
Harold Hamm didn’t come to America Innovates to defend fossil fuels. He came to defend reliability—the single trait that makes or breaks any infrastructure bet.
AI is the largest infrastructure investment since the internet. It will require more energy than anything we’ve built before. And that energy needs to be there 24/7.
If you’re a B2B leader, stop thinking of energy as a “sustainability checkbox.” Start thinking of it as a competitive advantage. The SaaS companies that understand where their compute comes from—and how to optimize for it—will win the next decade.
Hamm’s final words from the stage stayed with me:
“You can have a lot of great ideas. But without power, they’re just ideas.”
In 2026, power is the product. And oil and gas are the backbone.
Want to dive deeper? Check out the full America Innovates recording with Harold Hamm and Kerry Dolan. And if you’re building energy-adjacent SaaS, drop me a note. I’m watching this space closely.