While the world focused on GPU shortages, the real constraint on AI compute is now physical infrastructure. The bottleneck has moved to accessing power, building data centers, and finding specialized labor like electricians and acquiring basic materials like structural steel. Merely acquiring chips is no longer enough to scale.

The limiting factor for large-scale AI compute is no longer physical space but the availability of electrical power. As a result, the industry now sizes and discusses data center capacity and deals in terms of megawatts, reflecting the primary constraint on growth.

While data was once a major constraint for training AI, models can now effectively create their own synthetic data. This has shifted the critical choke points in the AI supply chain to physical infrastructure like power grids and data center construction, which are now the primary limiters of growth.

While AI training requires massive, centralized data centers, the growth of inference workloads is creating a need for a new architecture. This involves smaller (e.g., 5 megawatt), decentralized clusters located closer to users to reduce latency. This shift impacts everything from data center design to the software required to manage these distributed fleets.

Contrary to popular belief, the primary constraint on expanding AI infrastructure isn't GPU supply. It's the physical world: acquiring land, getting permits, and finding enough skilled tradesmen for construction and wiring. The GPUs are one of the last items to be installed in a long, labor-intensive process.

To secure the immense, stable power required for AI, tech companies are pursuing plans to co-locate hyperscale data centers with dedicated Small Modular Reactors (SMRs). These "nuclear computation hubs" create a private, reliable baseload power source, making the data center independent of the increasingly strained public electrical grid.

Satya Nadella clarifies that the primary constraint on scaling AI compute is not the availability of GPUs, but the lack of power and physical data center infrastructure ("warm shelves") to install them. This highlights a critical, often overlooked dependency in the AI race: energy and real estate development speed.

The race for compute power is moving from centralized data centers to decentralized networks. Companies are already putting GPU clusters next to homes, and Tesla is positioned to leverage its Powerwalls and Starlink for a distributed compute system that bypasses traditional infrastructure bottlenecks.

The infrastructure demands of AI have caused an exponential increase in data center scale. Two years ago, a 1-megawatt facility was considered a good size. Today, a large AI data center is a 1-gigawatt facility—a 1000-fold increase. This rapid escalation underscores the immense and expensive capital investment required to power AI.