Powering the AI boom: Investing in energy, infrastructure, and the data revolution

Rising power demand: The foundation of AI growth

AI computing runs hot. Training a large-scale model consumes as much electricity as several thousand homes. Each new model generation multiplies that demand. By the end of this decade, data centres could use close to 10% of total U.S. electricity.

Europe faces similar pressure. Ireland, the Netherlands, and Denmark have already limited new data centre projects to protect grid stability. In Asia, Singapore and South Korea are pushing through renewable projects to meet surging digital workloads.

These shifts create structural opportunity. Companies investing in grid expansion, renewable capacity, and power efficiency are moving to the centre of the digital economy. Utilities and infrastructure providers, often seen as defensive, are now growth assets.

Utilities and renewables: The energy backbone

Utilities such as NextEra Energy, Iberdrola, and RWE are scaling renewable capacity to serve data centres that run around the clock. Corporate clients like Microsoft and Amazon have signed long-term power purchase agreements, locking in green supply for the next decade. These contracts stabilise revenues and fund new solar and wind installations.

This is a shift from the traditional utility model. Steady demand from AI infrastructure offers consistent returns, supported by government incentives and private capital. Investors can gain access through renewable energy funds, infrastructure ETFs, or utilities with large clean-energy pipelines.

The same theme is playing out in Asia. Japan and South Korea are boosting offshore wind, while China leads in solar deployment. Global investors with a multi-region approach can capture the full buildout of the AI power grid.

Data centre REITs: The new real estate cycle

Data centres are the warehouses of the digital world. Demand from AI workloads is pushing them into a new expansion phase. There are some REITs benefiting from long-term leases indexed to power usage. Higher energy consumption means higher revenue per facility.

Construction costs have climbed, but so has pricing power. Tenants are paying premiums for reliable energy access and advanced cooling. Markets such as Northern Virginia, Singapore, and Frankfurt are leading global capacity growth.

For investors, data centre REITs provide exposure to tangible assets with recurring income. They also diversify portfolios heavy in software and semiconductors. REIT ETFs and infrastructure funds offer an accessible entry point for those seeking broad exposure.

The supply chain: Cooling and power efficiency

The energy challenge doesn’t end with generation. Efficient delivery and temperature control are essential. There are even companies that supply the systems that keep data centres running smoothly. Cooling alone accounts for roughly 40% of total energy use in a high-density facility.

Orders for liquid cooling and heat recycling technology are rising sharply as chip density increases. These industrial enablers have pricing power and multi-year order visibility. They offer cyclical exposure with real-world growth drivers beyond consumer demand.

Investors seeking diversified exposure can consider global infrastructure or industrial ETFs that hold these names. They balance the long-term energy trend with shorter innovation cycles in equipment manufacturing.

Energy security and policy tailwinds

Governments are now treating electricity as a strategic asset. In the U.S., new incentives under the Inflation Reduction Act are funding domestic clean-energy expansion. Europe’s REPowerEU plan aims to cut dependency on imported gas while supporting grid interconnection projects. Asia is fast-tracking battery storage and nuclear restarts.

This alignment between energy policy and digital growth suggests that infrastructure investment will stay elevated well into the next decade. Public funding, tax credits, and private capital are reinforcing the same trend: AI’s energy demands are too large to ignore.

Valuation and risk

Many renewable and infrastructure names have already outperformed broader markets, so entry points matter. Higher interest rates can pressure valuations for capital-intensive businesses. Investors should look for firms with secured power contracts, moderate leverage, and predictable cash flows.

Volatility in energy prices also affects margins, though long-term contracts often provide insulation. Blending growth and income exposure—utilities, REITs, and industrials—can smooth the risk profile.

How to build exposure

An AI infrastructure strategy can be layered across several parts of the market:

• Energy providers. Utilities and renewable producers expanding capacity for data centres.
• Infrastructure operators. REITs and construction firms developing and managing facilities.
• Industrial enablers. Companies supplying electrical systems, cooling, and automation.

Thematic ETFs focusing on clean energy, global infrastructure, or digital real estate capture this full value chain. Long-term investors may also consider infrastructure debt or green bonds for steady income exposure tied to the same underlying growth.

Long-term view: The physical side of digital growth

AI’s expansion depends on real assets—power plants, transmission lines, and warehouses of servers. These are capital-intensive, slow to build, and impossible to replace overnight. That makes them both scarce and valuable.

For investors, this part of the AI ecosystem offers growth supported by tangible demand and long-term contracts. It’s less volatile than software, yet essential to the same technological trend.

The digital economy will only scale as fast as its energy supply allows. The companies building that foundation—utilities, renewables, and data-centre operators—will be the quiet winners of the AI decade.