The race for AI supremacy is not just about models but about securing the underlying infrastructure. The most significant bottlenecks and price appreciation over the next 3-5 years will be in physical assets: land for data centers, permits to build, energy to power them, and the compute itself.

The AI buildout is unlikely to suffer a massive oversupply crash because it is constrained by real-world factors beyond chips: a lack of power, data centers, and even skilled trades like electricians. This acts as a natural governor, creating a longer, more durable investment cycle.

Government rules on where AI data centers can be built will concentrate development in specific geographies. This constraint creates a strong investment case for solutions that solve the resulting power and resource bottlenecks, such as on-site power generation, fuel cells, and energy storage systems.

The biggest investment opportunity lies in the beneficiaries of big tech's massive AI capital expenditures. This "food chain" includes data centers, power grid upgrades, and industrial suppliers who are seeing unprecedented demand for the foundational infrastructure AI requires.

Beyond being an inflation hedge, infrastructure represents a key constraint on AI's growth. Investing in areas like power capacity and data compute allows investors to "own the constraint on AI," providing a diversified way to gain exposure to the dominant technology theme.

The massive capital rush into AI infrastructure mirrors past tech cycles where excess capacity was built, leading to unprofitable projects. While large tech firms can absorb losses, the standalone projects and their supplier ecosystems (power, materials) are at risk if anticipated demand doesn't materialize.

The abstract race for AI superiority is now grounded in physical reality. Control over electricity grids, cooling, and land for data centers has become as strategically important as semiconductor supply chains, shaping who can scale frontier AI.

Unlike railroads or telecom, where infrastructure lasts for decades, the core of AI infrastructure—semiconductor chips—becomes obsolete every 3-4 years. This creates a cycle of massive, recurring capital expenditure to maintain data centers, fundamentally changing the long-term ROI calculation for the AI arms race.

The massive capital expenditure on AI infrastructure is not just a private sector trend; it's framed as an existential national security race against China's superior electricity generation capacity. This government backing makes it difficult to bet against and suggests the spending cycle is still in its early stages.