Despite widespread adoption, Patrick Collison notes that AI has not yet produced measurable gains in macroeconomic productivity. He points to recent studies and the lack of corresponding GDP growth outside the U.S. as evidence that the diffusion of these technologies through the economy is slow and complex.

The full economic impact of AI is constrained by the physical build-out of data centers. With only a quarter of the projected $3 trillion in necessary infrastructure capex deployed through 2028, widespread adoption and its labor market effects will be gradual, not instantaneous.

Even if AI progress stopped today, it would take 10-20 years for the economy to fully absorb and implement current capabilities. This growing gap between what's technologically possible and what's adopted in the market creates a massive, long-term opportunity for innovators.

Like past technological leaps, AI's economic impact will be sequenced. Expect immediate real income gains as new products emerge. The broader disinflationary effects from productivity improvements will only materialize later, after businesses fully re-engineer their operations.

Concerns about immediate AI-driven job losses are premature. True labor displacement requires a lengthy phase-in period for broad enterprise adoption, building new application layers, and integrating AI into existing workflows and processes, which takes significant time.

General-purpose technologies like AI initially suppress measured productivity as firms make unmeasured investments in new workflows and skills. Economist Erik Brynjolfsson argues recent data suggests we are past the trough of this "J-curve" and entering the "harvest phase" where productivity gains accelerate.

The AI productivity boom is confined to tech because developers have fewer adoption hurdles. Coding is a text-only medium with self-contained context in a codebase. In contrast, roles like marketing or law require complex data setup and workflow re-engineering, slowing down the productivity gains seen in macro-economic data.

Glenn Hutchins explains that broad economic efficiency gains from AI are not yet visible because companies are in the initial, costly investment phase. Meanwhile, they are just now reaping benefits from a decade of cloud investment, aided by a new generation of digital-native CEOs.

History shows a significant delay between tech investment and productivity gains—10 years for PCs, 5-6 for the internet. The current AI CapEx boom faces a similar risk. An 'AI wobble' may occur when impatient investors begin questioning the long-delayed returns.