By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel is bearish on Constellation Energy (CEG) due to uncertainty in AI power demand, competition from other power sources, and regulatory risks. They agree that CEG's nuclear fleet may not be the sole 'green' baseload solution for hyperscalers.
Risk: AI demand underwhelming or macro power prices swinging
Opportunity: Potential high-margin, long-term contracts with data-center customers
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
The artificial intelligence (AI) boom needs plenty of electricity.
Constellation Energy sits at the intersection of AI and energy.
The next AI bottleneck may be power, not computing.
For the past three years, investors have been obsessed with one question: Who will win the artificial intelligence (AI) race?
Most of the attention has gone to familiar names. Nvidia supplies the chips. OpenAI builds the models. Microsoft, Alphabet, Amazon, and Meta Platforms are spending billions of dollars building the infrastructure needed to power AI.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
At first glance, the race seems straightforward: Build better chips and models and bigger data centers.
But a surprising problem is emerging. Some of the world's largest technology companies are discovering that having enough AI chips may not be enough. They also need electricity -- a lot of it.
And that reality is quietly changing the economics of the entire AI industry.
One of the most important developments in AI over the past few years didn't come from a new chatbot or a breakthrough chip. It came from the energy sector.
In late 2024, Microsoft signed a long-term agreement tied to the restart of a nuclear reactor at Three Mile Island, a facility best known for the 1979 nuclear accident that changed public perception of nuclear energy in the United States.
One of the world's most valuable technology companies is helping bring a nuclear reactor back into operation. Why?
Because Microsoft has long understood something many investors are only beginning to appreciate. The future of AI depends on both computing power and electrical power.
Every AI interaction requires electricity. When you ask ChatGPT a question, generate an image, or use an AI-powered search engine, thousands of computers inside massive data centers work together to produce a response.
Those facilities consume enormous amounts of energy. As AI models become more capable and widely used, electricity demand continues to rise. That creates a challenge many investors never anticipated.
Building more chips and data centers is difficult. But building new power generation may be even harder. New power plants often require years of permitting, environmental reviews, regulatory approvals, and construction before they can begin producing electricity.
That's a problem because AI demand is growing today. Without power, even the most advanced AI chip becomes an expensive piece of silicon.
Most investors have probably never heard of Constellation Energy (NASDAQ: CEG). The company doesn't build AI models, manufacture semiconductors, or sell software. Instead, it is the largest producer of clean, reliable energy in the U.S., owning the largest fleet of nuclear power plants.
At first glance, that may not sound particularly exciting. But in today's environment, those assets may be becoming increasingly valuable.
Nuclear power provides something AI companies desperately need: large amounts of reliable, around-the-clock electricity. As data center demand grows, many experts believe the grid will require a mix of energy sources, including nuclear, natural gas, renewables, and battery storage.
Constellation's advantage is that it already owns one of the largest fleets of operating nuclear plants in the United States. This gives the company optionality as demand for electricity from data centers grow. For example, Constellation Energy could potentially sell electricity to higher-value customers, including data-center operators via direct long term agreements -- just like the agreement it did with Microsoft to supply electricity from Three Mile Island.
So, while most investors still view the company as a utility business, Constellation Energy sits at the intersection of two powerful trends: growing electricity demand and the rapid expansion of AI.
As owners of critical infrastructure often become some of the biggest winners during periods of major economic transformation, Constellation Energy could be that winner in this AI transformation.
In many ways, Constellation is more a bet that electricity will become one of the most valuable resources in the AI era than on nuclear power.
Whether Constellation Energy ultimately becomes a winning investment remains to be seen. The stock has already attracted significant investor attention, and expectations for future growth are much higher than they were a few years ago. That explains why the stock has risen by close to 600% in the last five years.
Still, investors should keep the stock under their radar. After all, one of the most important questions that AI investors should focus on for the next decade is simple: Who will supply the electricity?
In the United States, Constellation Energy remains the best-positioned company.
Before you buy stock in Constellation Energy, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Constellation Energy wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $443,191! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,258,838!
Now, it’s worth noting Stock Advisor’s total average return is 941% — a market-crushing outperformance compared to 211% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of June 6, 2026. *
Lawrence Nga has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Alphabet, Amazon, Constellation Energy, Meta Platforms, Microsoft, and Nvidia. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
Four leading AI models discuss this article
"Even with growing data-center demand, Constellation Energy faces structural headwinds—capital-intense nuclear assets, regulatory rate constraints, and competition from cheaper generation—making AI-tailwinds unlikely to deliver a sustainable, outsized re-rating."
The article casts Constellation Energy (CEG) as the AI electricity catalyst, but the bullish case rests on thin ice. AI power demand is real but uncertain in scale and timing; data centers can source electricity from diverse mix (gas, renewables, storage) and are highly price-sensitive. Nuclear expansion faces long permitting, high capital costs, and policy risk; even existing nuclear assets face operational and regulatory headwinds. CEG’s value would depend on favorable rate cases and long-duration PPAs with data-center customers—conditions that may not hold if competition intensifies or if project overruns occur. The stock’s 5-year run elevates valuation risk if AI demand underwhelms or if macro power prices swing.
The strongest counterpoint is that AI demand for power may be incremental rather than transformative, and competing generators (gas, wind/solar, storage) could meet much of that load at lower cost; plus, nuclear capital intensity and regulatory hurdles could cap upside for CEG, rendering the current optimism overextended.
"Constellation Energy has transitioned from a defensive utility into a cyclical growth stock tied to AI infrastructure, but current valuations leave little margin for operational error."
Constellation Energy (CEG) is effectively pivoting from a regulated utility model to a merchant power provider with high-margin, long-term contracts. The market is pricing in a structural shift where nuclear power becomes the 'baseload' for hyperscalers like Microsoft. However, at a forward P/E currently hovering near 30x, the stock is no longer a value play; it is a growth equity. The thesis relies on the assumption that data center power demand will outpace grid capacity indefinitely. While the Three Mile Island deal is a catalyst, investors must watch for regulatory pushback on rate hikes and potential cost overruns in nuclear maintenance, which could compress margins significantly.
The bull case ignores the massive capital expenditure required to upgrade aging nuclear infrastructure and the political risk that regulators may force utilities to prioritize residential grid stability over high-paying corporate data centers.
"CEG's power supply advantage is real but already priced in at 600% gains; the bigger risk is that data centers will diversify power sources and regulators will prevent utilities from capturing windfall AI-era margins."
The article conflates two separate theses: (1) AI will need massive power, which is true; (2) CEG is uniquely positioned to capture that value, which is far less certain. CEG trades at ~$100/share after a 600% five-year run—already pricing in significant upside. The Microsoft Three Mile Island deal is real but represents one contract, not proof of durable competitive moat. CEG faces headwinds: renewable + battery storage are accelerating faster than the article suggests, data centers are diversifying power sources geographically, and utility regulation caps returns on equity. The article also omits that CEG's fleet is aging and decommissioning risk is real.
If AI power demand truly becomes the binding constraint (not chip supply or capex), and if CEG can lock in 20+ year contracts at premium rates with tech giants, the stock's current valuation could be justified—especially if nuclear becomes politically rehabilitated.
"After a 600% rally, CEG already prices in much of the AI power thesis, leaving thin margin for the multi-year delays typical of nuclear restarts."
The article correctly flags power as the emerging AI constraint and CEG's nuclear fleet as a scarce asset, evidenced by the Microsoft-Three Mile Island restart deal. Yet it underplays restart timelines, which historically span 5-7 years amid NRC approvals, cost overruns, and public opposition. CEG shares have already climbed ~600% over five years, embedding aggressive growth assumptions. Direct PPAs to data centers offer margin upside, but competition from gas peakers and behind-the-meter renewables could cap pricing power. Investors must weigh whether incremental nuclear supply arrives fast enough to justify current multiples.
The Microsoft precedent could trigger a wave of similar high-value, long-term contracts that accelerate restarts and expand margins far beyond current utility benchmarks, making today's valuation appear conservative rather than stretched.
"Accelerated nuclear restarts are unlikely to unlock outsized margins; financing, regulation, and competition cap upside."
Grok's call that restarts could spur outsized margins hinges on a high-rate, near-term ramp that seems unlikely given NRC delays and public opposition; accelerating restarts would still face capex, scheduling, and cost overruns that squeeze ROE. Even with PPAs to data centers, the competitive risk from gas/renewables and potential rate-case pushbacks cap long-term pricing power. Risk-reward skewed toward downside if AI demand proves less sticky than assumed.
"CEG’s valuation reflects a scarcity premium for carbon-free baseload power that gas and renewables cannot replicate for hyperscalers."
Claude, you’re missing the regulatory 'moat' of nuclear: it’s not just about aging fleets, it’s about the impossibility of permitting *new* baseload capacity. While others worry about gas competition, they ignore that hyperscalers need 24/7 carbon-free energy to meet ESG mandates. Gas doesn't solve that. CEG isn't just selling power; they are selling the only viable 'green' baseload that satisfies both regulators and tech giants. The valuation isn't just growth; it's a scarcity premium.
"Nuclear's regulatory moat is real but doesn't guarantee pricing power if renewables + storage become cost-competitive for 24/7 carbon-free load."
Gemini's 'only viable green baseload' claim overstates CEG's moat. Microsoft's Three Mile Island deal proves demand exists, but hyperscalers can meet ESG mandates via long-term renewable PPAs + storage at lower cost than nuclear restarts. The regulatory barrier to *new* nuclear is real, but it doesn't protect existing assets from margin compression if renewables + batteries undercut pricing. CEG's scarcity premium assumes tech giants have no alternatives—they do.
"Hyperscalers' renewable-plus-storage diversification erodes CEG's claimed nuclear ESG exclusivity."
Gemini, your ESG moat claim ignores that hyperscalers like Microsoft already blend nuclear with massive renewable PPAs plus batteries and are siting data centers in regions rich in wind or hydro. If those alternatives scale faster than NRC restarts, CEG loses the 24/7 carbon-free scarcity premium that justifies 30x forward earnings. The Three Mile Island deal becomes an outlier rather than a template.
The panel is bearish on Constellation Energy (CEG) due to uncertainty in AI power demand, competition from other power sources, and regulatory risks. They agree that CEG's nuclear fleet may not be the sole 'green' baseload solution for hyperscalers.
Potential high-margin, long-term contracts with data-center customers
AI demand underwhelming or macro power prices swinging