By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel is divided on Constellation Energy's (CEG) potential as a data center power provider, with concerns about execution risks, regulatory hurdles, and political pushback outweighing the bullish case for merchant nuclear fleet upside.
Risk: The threat of regulatory intervention or windfall taxes due to residential rate hikes, as highlighted by Gemini.
Opportunity: The potential for uncontracted merchant capacity to capture higher realized power prices, as argued by Grok.
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 amount of electricity hyperscalers can reliably access is proving a critical bottleneck to their expansion plans.
Nuclear power companies and battery storage suppliers may both benefit from accelerating AI growth.
Constellation Energy could become a major power supplier to AI infrastructure companies.
Artificial intelligence isn't just a tech story. It's an electricity story, too.
Wall Street increasingly believes that the build-out of AI data centers could drive one of the largest expansions in power demand in decades.
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 »
According to Goldman Sachs, as hyperscalers race to build the infrastructure needed to operate increasingly complex AI models, global data center power demand could rise roughly 160% by 2030. Meanwhile, U.S. electricity demand from data centers alone may nearly triple during that period. Our power grids were not built for this kind of demand growth.
Large AI data centers consume enormous amounts of electricity. Some hyperscale campuses are now being designed to consume hundreds of megawatts of electricity, putting their power needs on par with those of medium-sized cities.
That is one reason some utility and power infrastructure stocks have been doing so well. Perhaps the most important one for investors to pay attention to is Constellation Energy (NASDAQ: CEG).
Constellation is currently the largest producer of nuclear energy in the U.S., operating a fleet of nuclear reactors capable of supplying stable baseload electricity around the clock.
The company also operates renewable energy and power management operations, including wind, solar, and hydroelectric power, as well as energy optimization services. That combination isn't trivial.
AI data centers increasingly require both reliable 24/7 electricity generation and broader grid flexibility to handle waxing and waning power demand. Nuclear plants can provide stable baseload electricity, while renewable energy sources paired with battery storage and grid-management systems are becoming increasingly important for balancing peak demand and stabilizing aging power grids.
In other words, Constellation is not simply a nuclear story.
However, nuclear energy is becoming central to the AI infrastructure discussion.
Microsoft recently signed a long-term commercial power agreement to restart one of the reactors at the former Three Mile Island nuclear facility to support future AI-related electricity demand. Amazon, Alphabet, and other hyperscalers have also continued investing heavily in nuclear, battery storage, and broader energy infrastructure projects.
Constellation generated about $4.2 billion in operating cash flow during 2025 while continuing to return capital to shareholders through dividends and share repurchases. Meanwhile, long-term contracted electricity demand tied to data centers and industrial customers continues increasing across multiple U.S. markets.
Of course, the company's valuation isn't particularly cheap anymore. Constellation stock rallied significantly over the past two years as investors increasingly recognized the connection between AI infrastructure and electricity demand. But the broader thesis may still have more room to run.
Indeed, the internet era created massive winners in fiber optics, semiconductors, cloud computing, and wireless infrastructure. AI may now be creating a similar investment cycle centered around electricity generation, transmission infrastructure, cooling systems, battery storage, and grid modernization.
That does not mean that investing in the sector is risk-free, though. Regulatory headwinds, power pricing challenges, political risk, and interest rates remain legitimate concerns.
But if AI infrastructure spending continues to accelerate, electricity producers with scalable generation capacity, renewable infrastructure, and reliable long-term power delivery may increasingly become some of the most strategically valuable assets in the market. And among them, Constellation stands out as the gold standard.
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 $477,813! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,320,088!
Now, it’s worth noting Stock Advisor’s total average return is 986% — a market-crushing outperformance compared to 208% 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 May 25, 2026. *
Jeff Siegel has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Alphabet, Amazon, Constellation Energy, Goldman Sachs Group, and Microsoft. 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
"Regulatory and interconnection delays will likely push most incremental nuclear cash flows from AI contracts into the late 2020s, muting near-term re-rating potential despite the demand narrative."
The article correctly flags surging data-center power demand as a structural tailwind, with Goldman Sachs projecting U.S. data-center load nearly tripling by 2030. Constellation Energy's 21 GW nuclear fleet offers the 24/7 carbon-free baseload that hyperscalers are now contracting directly, as shown by the Microsoft-Three Mile Island restart. Yet the piece underplays execution risk: NRC relicensing and grid interconnection queues routinely stretch 4-7 years, while already-elevated valuation (post-2023 rally) leaves little margin if power prices or capacity factors disappoint. Renewables-plus-storage competition and potential federal subsidy changes add further uncertainty to the multi-year thesis.
Microsoft's TMI deal demonstrates regulators can accelerate approvals when national AI competitiveness is at stake, potentially compressing timelines and validating the bullish case faster than historical nuclear precedents suggest.
"The power-demand thesis is sound, but CEG's valuation and the structural shift toward hyperscaler self-supply mean the stock likely offers lower risk-adjusted returns than the article implies."
The article conflates two separate theses without stress-testing either. Yes, AI data centers will consume massive power—Goldman's 160% global demand growth by 2030 is plausible. But CEG's valuation has already priced in much of this. The real risk: hyperscalers are increasingly self-building nuclear (Microsoft–Three Mile Island) and signing PPAs (power purchase agreements) at fixed, often below-market rates. This locks in revenue but caps upside and shifts negotiating leverage away from utilities. CEG's 2025 cash flow ($4.2B) is solid, but the article doesn't address whether contracted data-center power actually expands margins or merely locks in volume at compressed spreads. The 'gold standard' framing obscures that CEG faces regulatory, permitting, and grid-interconnection delays that could defer revenue realization by years.
If hyperscalers continue building captive nuclear and battery capacity while locking utilities into long-term fixed-price contracts, CEG's growth becomes a margin squeeze disguised as volume growth—and the stock's 2-year rally may have already priced in the best-case scenario.
"The market is incorrectly applying software-like valuation multiples to a regulated utility asset class that remains vulnerable to political and regulatory price-capping."
Constellation Energy (CEG) is being priced as a high-growth tech utility, but investors are ignoring the 'utility trap.' While the AI power demand surge is real, utilities are heavily regulated. Even with lucrative data center contracts, rate-of-return caps often prevent utilities from capturing the full 'AI premium' that software companies enjoy. CEG is currently trading at a significant premium to its historical P/E, banking on a massive expansion of nuclear baseload that faces immense regulatory and NIMBY-related friction. If the AI build-out hits a cooling or efficiency bottleneck, or if regulators force lower pricing to protect residential consumers, CEG’s multiple will compress rapidly. This is a bet on infrastructure, not software margins.
The counter-argument is that hyperscalers like Microsoft and Amazon are willing to bypass traditional utility rate structures through direct-connect private power agreements, effectively turning CEG into a merchant power provider with pricing power that traditional utilities never had.
"The AI electricity boom is real, but Constellation Energy faces valuation, capex, and policy/regulatory risks that could limit near-term upside even if AI demand grows long term."
AI data centers are likely to lift power demand, but the article paints a one-way picture. Goldman’s scenario of 160% global growth by 2030 hinges on sustained capital availability, PPAs, and grid upgrades that may stall. For Constellation Energy, the pitch rests on a dual role as nuclear baseload plus renewables/storage, yet policy risk, rate-base hurdles, and large capex could compress near‑term returns even if long‑term demand grows. CE generated about $4.2 billion in operating cash flow in 2025, and its multiple expansion depends on a favorable electricity-price and regulation backdrop. The “gold standard” label may gloss valuation and execution risks specific to nuclear restart timelines.
The AI-power boom could prove slower or less grid-friendly than advertised, with efficiency gains dampening megawatt growth; CE’s regulatory and capex headwinds could keep returns muted even if demand expands.
"CEG's merchant exposure lets wholesale prices lift uncontracted output beyond fixed-PPA constraints."
Claude flags PPAs shifting leverage and capping spreads, yet this underweights CEG's merchant nuclear fleet, where a material share of output clears into wholesale markets already seeing data-center-driven price spikes. If Grok's 4-7 year interconnection delays materialize, uncontracted capacity could capture higher realized power prices than any fixed PPA, offsetting the margin-squeeze risk rather than confirming it.
"CEG's merchant upside assumes scarcity; if hyperscalers self-build nuclear, CEG shifts from anchor to marginal supplier, collapsing realized prices."
Grok's merchant-fleet upside is real but assumes uncontracted capacity persists through 2030. Claude's PPA-lock thesis overlooks that CEG's existing fleet (21 GW) already operates at high capacity factors; new contracts don't displace merchant revenue—they layer on top. The actual squeeze risk: if hyperscalers build captive nuclear faster than CEG adds capacity, CEG becomes a marginal supplier filling gaps at whatever price clears, not a baseload anchor. That's deflationary for realized spreads, not inflationary.
"Regulatory intervention to protect residential ratepayers from AI-driven price spikes is a greater threat to CEG's valuation than PPA margin compression."
Claude and Grok are debating merchant versus PPA pricing, but both ignore the political economy of the 'utility trap' Gemini mentioned. If data centers cause residential rate hikes, state PUCs will force CEG to prioritize grid stability over merchant profits or hyperscaler contracts. The risk isn't just margin compression; it's the threat of windfall taxes or mandated price caps on nuclear output. CEG’s political license to operate is the most undervalued variable in this entire thesis.
"Merchant upside is fragile due to regulatory/margin caps and transmission constraints, not robust uncontracted price spikes."
Grok, your merchant upside hinges on uncontracted output clearing into wholesale markets at high prices, but that path is narrow. Regulatory and PUCs can cap merchant margins to protect consumers; transmission constraints and regional price layering will suppress spikes; and rising capex and financing costs may force CEG to rely more on regulated returns, not merchant windfalls. In short, pure merchant upside may be the weakest link in the CEG thesis.
The panel is divided on Constellation Energy's (CEG) potential as a data center power provider, with concerns about execution risks, regulatory hurdles, and political pushback outweighing the bullish case for merchant nuclear fleet upside.
The potential for uncontracted merchant capacity to capture higher realized power prices, as argued by Grok.
The threat of regulatory intervention or windfall taxes due to residential rate hikes, as highlighted by Gemini.