National Grid has increasingly found itself mentioned alongside discussions about artificial intelligence, data centres, and the future of electricity demand. As investors search for companies that could benefit from the rapid expansion of AI infrastructure, transmission operators have naturally entered the conversation. The logic appears simple. Artificial intelligence requires vast amounts of electricity. Electricity requires grids. National Grid owns critical parts of those grids.
While that conclusion is not entirely wrong, it often leads investors to misunderstand what National Grid actually is and how it creates shareholder value.
Many investors, particularly those in the United States, approach the company through the lens of a growth stock. They see rising electricity demand and assume profits should rise accordingly. This is how many industries operate. More customers typically mean more sales, higher margins, and potentially faster earnings growth. National Grid operates under a very different model.
National Grid is first and foremost a UK regulated infrastructure business. Its earnings are largely determined by the assets it owns and the returns regulators allow it to earn on those assets. If electricity demand were to double tomorrow, National Grid would not simply double its profits. Instead, the company would need to propose new investments, receive regulatory approval, construct the infrastructure, and then earn a regulated return on those investments over many years.
Understanding this is critical because it changes the nature of the investment case entirely.
Artificial intelligence does not transform National Grid into a technology company. It creates another source of demand for future infrastructure investment. Data centres require transmission capacity. Transmission capacity requires grid upgrades. Grid upgrades increase the regulated asset base. A larger regulated asset base supports higher earnings over time.
The process is slower than many investors expect, but it is also far more predictable.
This was evident in National Grid’s latest results. The company invested £11.6 billion during the year and expects investment to increase to approximately £13 billion next year. Management is targeting regulated asset growth of around 10% annually through 2031 while expecting underlying earnings per share growth of 8% to 10% annually.
For investors accustomed to software companies and semiconductor manufacturers, these growth rates may not appear particularly exciting. For infrastructure investors, they are remarkably attractive. The difference lies in the visibility of those returns. National Grid is not attempting to create a new market or launch a disruptive product. It is building essential infrastructure that governments, businesses, and consumers increasingly depend upon.
The company’s growth is also supported by multiple long-term trends rather than a single narrative. Electrification continues to expand across transportation and industry. Renewable energy requires significant investment in transmission networks. Energy security has become a strategic priority across Europe and North America. Data centres are consuming increasing amounts of electricity. Each of these developments creates additional demand for grid investment.
This is why National Grid should not be viewed as an AI stock.
Artificial intelligence is simply one of several forces increasing demand for electricity infrastructure. The company benefits not because AI itself is profitable, but because AI contributes to the need for new transmission lines, substations, interconnectors, and network upgrades. Once these assets are built and incorporated into the regulated asset base, they can generate returns for decades.
Recent discussions surrounding a regulatory dispute in New England provide a useful example of how investors can sometimes overreact to headlines. The issue relates to the level of return allowed on a specific portion of National Grid’s transmission assets. While this may create a modest headwind, it does not alter the fundamental business model. The company still owns the assets. The assets still generate revenue. The debate concerns the size of the permitted return rather than the existence of the return itself.
More importantly, National Grid’s investment thesis does not depend on a single regulatory decision. The company operates across multiple jurisdictions, including both the United Kingdom and the United States. Investors are not buying National Grid because of one transmission project in New England. They are investing in a diversified portfolio of regulated infrastructure assets with decades of investment opportunities ahead.
This is perhaps the most important point for investors to understand.
National Grid is unlikely to deliver the explosive returns associated with many AI-related stocks. Regulators would never allow the economics necessary to support those types of outcomes. However, the company does not need to become the next Nvidia to be a successful investment.
What National Grid offers is something different. It offers long-term asset growth, predictable earnings, inflation-linked cash flows, and a growing dividend supported by infrastructure that modern economies cannot function without.
In many ways, this makes the company one of the more misunderstood beneficiaries of the AI era. The market often focuses on the technologies that consume electricity while overlooking the networks that deliver it. Yet without transmission infrastructure, none of those technologies can operate at scale.
For investors seeking the next AI winner, National Grid may appear too slow and too regulated. For investors seeking a durable infrastructure compounder benefiting from the same forces driving AI adoption, it may be exactly the type of business worth understanding.