Energy companies are becoming the unlikely beneficiaries of the AI boom, with firms securing long-term power contracts and nuclear partnerships positioning themselves for decades of stable revenue. While GPU makers and chip designers grab headlines, the companies that can reliably deliver electricity to power-hungry AI infrastructure are attracting serious investor attention and analyst upgrades.

Why Is Power Suddenly the Bottleneck in AI Infrastructure?

The explosive growth of AI data centers has created an energy crisis that's reshaping the technology landscape. Companies building massive AI infrastructure need not just computing power, but reliable, abundant electricity to keep their systems running 24/7. This has shifted focus from traditional tech winners to the unglamorous but essential business of power generation and distribution.

The challenge isn't just about having enough electricity; it's about securing it reliably for decades. AI hyperscalers, the massive companies building data centers to train and run artificial intelligence models, are signing long-term power purchase agreements that lock in energy supplies for 20 years or more. This stability is worth billions in future cash flow, making energy companies attractive to investors who previously overlooked them.

Which Energy Companies Are Winning the AI Power Race?

Constellation Energy Corporation stands out as a major player in this shift. The company has secured regulatory approval to restart the Three Mile Island nuclear power plant, a facility that will serve Microsoft's data center energy needs through a 20-year agreement. This isn't a small deal; the company also holds a similar 20-year power purchase agreement with Meta Platforms. Management expects Constellation to generate more than $20 billion in free cash flow through the end of 2029, with lower regulatory hurdles helping accelerate that timeline.

Wells Fargo reaffirmed its price target on Constellation Energy at $516, the highest target on Wall Street, signaling confidence in the company's position as a critical infrastructure provider for the AI era. The regulatory progress on restarting Three Mile Island is particularly significant because it demonstrates that policymakers recognize the urgency of solving AI's power problem.

How Are Data Center Companies Addressing Power Constraints?

Data center operators themselves are also becoming attractive investments as they navigate the power challenge. Applied Digital Corporation has managed to secure power access and deliver projects on time, advantages that set it apart in a competitive market. On June 8, the company announced a new long-term lease at its Delta Forge 2 site with a US-based high-investment-grade hyperscaler, marking its fifth AI data center campus overall.

B. Riley raised Applied Digital's price target from $53 to $66 in late May, implying 59% upside from current levels. The analyst firm noted that cloud companies are signing deals faster than ever, and investors are placing greater value on firms that have secure access to power and can deliver projects on time. Local community opposition, grid interconnection delays, and electrical equipment shortages have become real obstacles for new developments, making companies that can overcome these challenges especially valuable.

IREN Ltd, an Australian data center operator, is also expanding aggressively. On June 3, the company reported signing a transmission connection agreement for an 800-megawatt data center in South Australia, its first facility in Australia. The company has also partnered with NVIDIA to deploy up to 5 gigawatts of AI infrastructure across its global data center pipeline, with NVIDIA even taking an option to purchase up to 30 million IREN shares over five years at $70 per share, potentially investing up to $2.1 billion.

What Makes These Companies Different From Traditional Tech Stocks?

The appeal of energy and infrastructure companies in the AI era comes down to predictability and scale. Unlike software companies that face rapid disruption, or chip makers dealing with supply chain volatility, energy companies with long-term contracts have visibility into revenue for decades. This stability is exactly what investors seek when betting on infrastructure plays.

• Long-Term Contracts: Energy companies like Constellation are locking in 20-year agreements with major AI companies, providing decades of predictable revenue streams that reduce business uncertainty.
• Nuclear Advantage: Companies with access to nuclear power, like Constellation with Three Mile Island, offer carbon-free, reliable baseload power that can run continuously without the intermittency issues of renewable sources.
• Regulatory Tailwinds: Government support for restarting nuclear plants and upgrading grid infrastructure is accelerating, removing barriers that previously slowed energy company growth and profitability.
• Scarcity of Solutions: Unlike GPUs or chips where multiple manufacturers compete, power infrastructure is geographically constrained and takes years to build, making existing capacity and connections extremely valuable.

The analyst community is catching on to this shift. Chen Goldberg, executive vice president of Product and Engineering at CoreWeave, a cloud infrastructure company, noted that enterprises deploying AI agents in production are accelerating their path toward more capable systems. This acceleration directly translates to higher power consumption and greater demand for the energy companies that can supply it.

What Should Investors Know About This Trend?

The AI power crisis is not a temporary problem that will resolve itself. As AI models grow larger and more companies deploy AI infrastructure, electricity demand will only increase. This creates a multi-decade tailwind for companies that can reliably deliver power to data centers. Unlike many technology trends that fade or shift, the need for electricity is fundamental and growing.

However, not all energy companies are equally positioned. Those with existing nuclear assets, regulatory approval for expansion, or long-term contracts with major hyperscalers have significant advantages. Companies struggling with grid interconnection or lacking power purchase agreements face headwinds that could limit their ability to capitalize on the AI boom.

The shift in investor focus from chip makers to energy companies reflects a maturing understanding of AI infrastructure. The bottleneck isn't computing power anymore; it's the electricity to run that computing power. This fundamental reality is reshaping which companies will be the real winners of the AI era, and energy companies are finally getting the recognition their critical role deserves.