The Great Inversion: AI Transforms Scarcity Into Abundance in Work

Estimated reading time: 6 minutes

AI did not devalue expertise. It inverted the scarcity map. When the output is free, the cost of selection becomes the premium, and the expensive part is everything we assumed professionals already had.

For most of the knowledge economy’s history, the thing that was hard to produce was the thing you got paid to produce. A strategy deck took days. A comprehensive brief required judgment so thoroughly embedded in format and syntax that the format became indistinguishable from the judgment. The artifact was the proof. The production cost kept the proof credible. That economy is gone. What replaced it is, I think, a clarification.

The Map That Was

Peter Block observed, writing about consulting, that the core transaction of any professional engagement is the transfer of expertise from the practitioner to the client. For most of professional history, that transfer happened through what you produced. You did not ship your judgment. You shipped a deck, a model, a memo. The artifact carried the expertise, made it portable and legible.

This is not a new pattern. The economic historian David Landes, writing about the first industrial revolution, described its central rupture as “a certain separation of the producer from production, an orientation to the market instead of to the shop.” The craftsperson was pulled from their craft. What the craft signaled became separable from what the craftsperson knew (See: Learning to Unlearn). We are watching the same separation happen again, in a different industry, at a different speed.

What AI collapsed was the production cost. The expertise behind the production remained intact. These two things were bundled so tightly for so long that the distinction barely seemed worth making.

It matters enormously now.

The Inversion

Klaus Schwab observed that in the digital economy, “a unit of wealth is created today with much fewer workers… because digital businesses have marginal costs that tend towards zero.” AI applied that same logic to cognitive production. The deliverable, the deck, the brief, the analysis, became a near-zero-marginal-cost good. The entire pricing architecture built on its scarcity had to follow.

Daniel Kahneman describes professional judgment as occupying a space “between questions of fact or computation on the one hand and matters of taste or opinion on the other.” AI is extraordinary at the computational end. What it cannot do is navigate the middle ground Kahneman describes, hold context in both hands and decide what to do with it. It cannot be accountable for its conclusions. A major MIT NANDA study found that when organizations deploy AI without redesigning how judgment is incorporated into the work, 95% of those efforts produce no measurable business value. The deliverable is free. Judgment is the bottleneck precisely because AI has no internal mechanism for doubt; it provides a smooth surface of certainty even when the underlying terrain is shifting and ambiguous.

The scarcity map inverted. What was rare, quality production,is now abundant. What was assumed, the judgment, taste, and accountability underneath the production, is now the expensive part.

What Is Now Expensive

Ana Andjelic, writing about the modern aspiration economy, observed that aspiration has shifted: it is now about having the taste to know what to choose, rather than simply the means to acquire anything. She was writing about consumers. The principle maps directly to professional work.

Taste is now expensive. This kind of calibrated judgment isn’t found in a manual; it’s the result of the kind of apprenticeship that prioritizes real-world problem solving over legacy credentials. In the professional sense: the calibrated capacity to know which option is right for this situation, this audience, this moment. Taking fifty AI-generated outputs and identifying the three worth using, and being able to explain the selection with enough specificity that the explanation is itself a contribution, that is expensive. Accountability is expensive. Being the person who stands behind the decision, who can be found when the decision turns out to have been wrong, is rarer than the professional world has typically acknowledged.

Here is where the trap lives. If production is free, taste risks becoming performative. There is a real difference between curation and the simulation of curation, between a professional who selects from genuine expertise and one who uses AI to mimic the appearance of selection. The distinction cannot be faked over time. It shows up in the reasoning, not in the output. Which is precisely why The New Legibility demands that the reasoning become visible.

What the Inversion Revealed

Here is the counterintuitive claim most AI-and-work conversations skip past: the inversion is a revelation. The economy it exposed was always there.

In 1927, Sherwood Anderson wrote a letter to his teenage son: “Next to occupation is the building up of good taste. That is difficult, slow work. Few achieve it. It means all the difference in the world in the end.” He was writing before AI existed. He was describing the economy we just arrived at.

The scholar D. Graham Burnett put it more directly: “The real project was always us: the work of understanding, and not the accumulation of facts… They are the work of being, not knowing, and knowing alone is utterly unequal to the task.”

Production was the visible layer. Being, the accumulated judgment, the calibrated taste, the willingness to stake your name on a position, was always underneath it. The inversion stripped the visible layer away. What remains is what was always the point.

The New Legibility

The demand the inversion places on professionals is a challenge of legibility, not a challenge to capability. The underlying skills always mattered. What the new economy requires is that they become visible.

Consider what this looks like in practice. In the old map, a consultant delivered a forty-page strategy deck. In the inverted map, the consultant shows the three analytical frames they considered and set aside, names the single assumption the whole recommendation rests on, and flags the one data point the model keeps ignoring, explaining why that outlier is actually the center of the strategy. The deck is the commodity. The reasoning is the product.

Judgment has to be shown, not just exercised. Taste has to be made explicit: the alternatives considered, the criteria used to choose, the confidence level attached to the recommendation. Accountability has to be staked before the outcome arrives, not claimed afterward.

Showing your work is how expertise proves itself, requiring a mindset of Perpetual Beta where identity remains fluid and the learning loop never closes.

The altitude on the scarcity map changed. The terrain held. The professionals who understood the terrain are exactly as valuable as they always were. Those who knew only the old map are discovering that it does not lead anywhere reliable anymore.

This continues the thread from The Proof of Work Problem, where we explored how AI collapsed the output-based signaling infrastructure of professional careers, and what new proofs of expertise the new economy demands.

Forward this to: senior individual contributors, strategic advisors, and leaders managing talented teams who are struggling to articulate why their expertise still commands a premium when AI can produce their deliverables on demand.

A Question for You: When AI can generate the output, what does a colleague need to show you to convince you the judgment behind it was genuinely theirs?

Madam I’m Adam