Stored Pressure

In the AI Story

The semiconductor analogy illustrates the mechanism. The first transistor (1947) replaced the vacuum tube and satisfied existing needs for electronic switching. Adoption was significant but measured. The integrated circuit (late 1950s) satisfied needs the transistor had revealed but could not meet — smaller circuits, cheaper manufacturing. Adoption was faster because pressure had accumulated since the transistor revealed what was possible but could not deliver what was needed. The microprocessor (1971) released a decade of accumulated pressure; adoption was faster still. Each technology in the sequence both satisfies pressure and creates it.

Cost determines when pressure can release. A technology does not release stored pressure simply by existing; it releases pressure by becoming cheap enough that the people experiencing the pressure can access it. Large language models existed before ChatGPT — GPT-3 was available through an API in 2020 — but the cost in dollars, technical expertise, and organizational overhead confined the technology to a small population. When OpenAI offered ChatGPT as a free consumer product, the cost dropped to zero and decades of stored pressure for a natural-language interface to computation discharged in two months.

The hundred-dollar-per-month price point of Claude Code represents the next stage: release of professional building pressure that had been accumulating for decades. Developers had spent careers learning languages they would have preferred to speak, debugging errors they would have preferred to describe. The stored pressure framework explains why modest technologies at the right price point outperform brilliant technologies at the wrong one, and why each subsequent price reduction will produce an adoption event faster than the last — because stored pressure compounds alongside cost reduction, and the product of the two determines the release rate.

The framework extends to populations not yet participating in the AI economy. Every year that a teacher in rural India cannot access a personalized curriculum generator, pressure increases. Every year that a small-business owner in Nairobi cannot afford custom inventory software, pressure increases. When the cost curve crosses the thresholds these populations occupy, the release will be proportionally larger — and faster — than what ChatGPT produced.

Origin

The stored pressure model is articulated in this volume as a synthesis of Edo Segal's observations in The Orange Pill about adoption-curve acceleration and Moore's cost-centric framework for understanding when capability reaches users. The physical metaphor — potential energy accumulating against a dam until a release mechanism lowers the barrier — draws on thermodynamic intuitions common to engineering culture but has not, to the author's knowledge, been formalized elsewhere.

Key Ideas

Adoption speed measures accumulated need, not product quality. The two-month adoption of ChatGPT was a measure of how long the dam had held, not of how good the technology was.

Each technology both satisfies and creates pressure. Satisfaction is immediate and visible; pressure creation is gradual and invisible, accumulating in the background until the next release mechanism arrives.

Cost determines release timing. Brilliant technology at the wrong price releases no pressure; modest technology at the right price can release decades of accumulated need.

Pressure compounds alongside cost reduction. The product of accumulated pressure and cost reduction determines release rate — and both are growing.

The AI musical greeting card has not yet been built. The most consequential applications of AI, by Moore's precedent, will be ones current researchers would dismiss as trivial until the cost threshold that reveals them is crossed.