The standard for measuring large compute deals has shifted from number of GPUs to gigawatts of power. This provides a normalized, apples-to-apples comparison across different chip generations and manufacturers, acknowledging that energy is the primary bottleneck for building AI data centers.

Analyst Chris Miller argues China's core challenge is manufacturing, as it lacks the advanced lithography tools monopolized by ASML. The US and Taiwan are projected to produce 30 times more quality-adjusted AI chips, a gap unlikely to close soon.

A single year of Nvidia's revenue is greater than the last 25 years of R&D and capex from the top five semiconductor equipment companies combined. This suggests a massive 'capex overhang,' meaning the primary bottleneck for AI compute isn't the ability to build fabs, but the financial arrangements to de-risk their construction.

The AI industry's growth constraint is a swinging pendulum. While power and data center space are the current bottlenecks (2024-25), the energy supply chain is diverse. By 2027, the bottleneck will revert to semiconductor manufacturing, as leading-edge fab capacity (e.g., TSMC, HBM memory) is highly concentrated and takes years to expand.

The primary constraint on AI scaling isn't just semiconductor fabrication capacity. It's a series of dependent bottlenecks, from TSMC's fabs to the limited number of EUV machines from ASML, and even further down to ASML's own specialized suppliers for components like lenses and glass.

The critical constraint on AI and future computing is not energy consumption but access to leading-edge semiconductor fabrication capacity. With data centers already consuming over 50% of advanced fab output, consumer hardware like gaming PCs will be priced out, accelerating a fundamental shift where personal devices become mere terminals for cloud-based workloads.

The 2024-2026 AI bottleneck is power and data centers, but the energy industry is adapting with diverse solutions. By 2027, the constraint will revert to semiconductor manufacturing, as leading-edge fab capacity is highly concentrated and takes years to expand.

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.

The long-term ability to scale AI compute is not constrained by power or data centers, but by the production of advanced semiconductors. The ultimate chokepoint is ASML, the world's only manufacturer of EUV lithography tools, which can only produce just over 100 units annually by 2030.