From Distribution to Cognition

The internet era made reach nearly free. The AI era is making high-quality thought nearly free. These are not the same shift — and confusing them is the most expensive strategic mistake a company can make right now.

Two eras, one axis

The internet did not invent connectivity. ARPANET, TCP/IP, and electronic mail all preceded the web. What the internet did was make the cost of distributing information — reaching anyone, anywhere, instantly — collapse toward zero. That single economic shift restructured every industry it touched.

Distribution leverage was the alpha of that era. The companies that won did not necessarily have the best products. They won because they mastered the movement of attention, goods, and information at scale. Logistics networks, search indices, social graphs, marketplace rails — all of them were fundamentally engines of distribution.

Artificial intelligence is doing something structurally different. It is not compressing the cost of reaching people. It is compressing the cost of thinking — of reasoning, synthesising, analysing, and generating structured recommendations. This is cognition leverage, and it operates on an entirely different economic axis.

The electricity parallel

Electricity was a general-purpose technology that amplified physical labour. It did not replace the factory — it restructured what the factory could do and who could do it economically. The companies that understood electricity as infrastructure, not novelty, built the industrial century.

AI is the general-purpose technology that amplifies mental labour. The companies that treat it as a productivity tool — a faster way to do what they already do — will gain marginal efficiency. The companies that redesign their cognitive architecture around it will build something structurally different: organisations that think faster, decide better, and allocate capital more precisely than any human team could alone.

We are approximately at 1885 in the electricity analogy. The infrastructure exists. The applications are nascent. The restructuring has barely begun.

The pocket strategist

One downstream implication crystallises the shift. Before AI, access to Tier-1 strategic thinking — the kind that structures ambiguous problems, runs rapid scenario analysis, and delivers prioritised recommendations — was radically scarce and expensive. It was rationed to organisations large enough to engage a major strategy firm at significant cost.

That scarcity is collapsing. A well-deployed AI system can today deliver structured strategic analysis, competitive teardowns, go-to-market frameworks, and decision packages at near-zero marginal cost. The pocket strategist is real.

But the analogy runs deeper than access. Consider precisely what a major strategy engagement delivers. The consultants do not primarily sell brilliant ideas. They sell compressed time-to-insight and de-risked decision-making. They absorb the mess — ambiguous mandates, political complexity, sprawling data — run the heavy cognitive process internally, and return a clean, prioritised set of action directives. The client receives output, not process.

This is the model AI must be held to. Not a conversational partner for open-ended exploration. An engine that ingests messy inputs, does the heavy lifting internally, and returns executable outputs with minimal demand on the user's time.