The AI Supercycle: Rewiring the Global Economic Triangle

We are currently living through what many experts call a "supercycle", a massive, structural shift in the global economy that is arguably bigger than the internet, mobile, and cloud revolutions combined. But while the excitement is palpable, the economic reality is currently shaped like an inverted triangle.

The Laws of Physics: Why the Triangle is Inverted. In the traditional software world, marginal costs were near zero; once you built the code, distributing it to millions was virtually free. AI breaks this model. Every time a user interacts with a model, you have to "burn GPUs" meaning the incremental cost of an AI user is significantly higher than that of a traditional software user.

Currently, the vast majority of the economic value in AI is trapped at the base of the stack, in semiconductors like Nvidia, while the application layer at the top remains relatively small. For this triangle to "flip" into a sustainable ecosystem, we must move from a period of massive capital expenditure (laying the "railroads" of AI) to a period of widespread economic value creation.

Infrastructure: The Quest for "Electrons to Tokens". The physical manifestation of this boom is found in the data center. We have moved into an era where the primary bottleneck is no longer just the chips themselves, but access to "energized" data centers. Projects like the 2.1-gigawatt campus in Abilene, Texas—large enough to power two cities the size of Denver—represent the "Manhattan Project" scale of current investments.

The goal for infrastructure providers is simple yet profound: converting electrons into tokens. As we move from training massive models to inference-time reasoning, the demand for these tokens is expected to grow not by 10x or 100x but by a staggering billion-fold.

The "Context Gap": Why Productivity Gains Take Time. Despite the power of these models, many enterprises are still waiting for a "productivity miracle". This is due to the "context gap". While we may already have AGI (Artificial General Intelligence) by many definitions, these models often lack the specific institutional knowledge—the "John or Jane" who has been at the company for 20 years and knows exactly how things work.

History shows us that it took 40 years for the electric engine to show productivity gains because factories had to be entirely rewired from their old steam-powered layouts. Similarly, the AI revolution requires a human refactoring problem: we must rewire our business processes to feed institutional context into the silicon.

Your Strategy: Becoming "Bionic". As the supercycle unfolds, the advice for individuals and leaders is clear: make yourself bionic. While raw IQ is becoming a commodity that machines can handle, EQ (emotional intelligence), human networks, and the ability to persuade and lead are becoming more valuable than ever.

The "SAS apocalypse" may disrupt software that hasn't innovated in a decade, but it creates a massive opportunity for those who can leverage AI to deliver "abnormal value".

A Final Word: Chill Out and Look Long-Term. It is easy to get caught up in the "fear-mongering" or the frantic pace of Twitter trends. However, the most successful strategies will be those that make secular bets on long-term shifts. Whether it’s reinventing healthcare, education, or the electrical stack itself, the value will ultimately accrue to those who can bridge the gap between raw compute and real-world utility.

The "laws of physics" for AI are still being written. The goal isn't just to watch the supercycle, it's to figure out where you fit into the new triangle.