Not Enough Problems

Solutions are getting cheaper. Problems are getting scarce—not because the world lacks conditions worth addressing, but because we are getting better at answering questions faster than we are getting better at framing them.

In 2016, Washington State put Initiative 732 on the ballot — a revenue-neutral carbon tax designed to be as economically rational as climate policy gets. Cut the sales tax, offset it with a price on carbon. The science was undisputed. The economics were broadly agreed upon. The initiative lost 59-41.

The coalition that killed it came from within the climate movement. Environmental justice organizations and clean energy alliances organized against the measure over what the revenue would mean. They wanted carbon revenues invested in frontline communities rather than returned as broad-based tax cuts. The two sides agreed on the mechanism — put a price on carbon — and couldn’t resolve what fairness required. No data settles that. No model adjudicates between distributive justice and aggregate welfare. The coalition had not settled the values its policy would serve. Agreement on the mechanism was not enough to carry the proposal.

A problem is an interpretation. It’s a claim that some gap between what is and what ought to be deserves organized effort. Loneliness is a problem only if you’ve committed to the value of human connection. Soil degradation is a problem only if you hold commitments about intergenerational responsibility. Each becomes a problem when someone commits to a value it threatens. Once that framing work is done, each opens further questions that AI could help answer.

Hayek’s great insight was that markets coordinate factual questions without requiring participants to understand the whole picture. A shipper doesn’t need to know why tin prices rose; the price signal tells her to economize. AI extends this capacity enormously. Code that took teams weeks, analysis that demanded expensive expertise — domain after domain, the constraints that justified elaborate organizational structures are dissolving.

His formulation leaves a prior question open. The price system enables actors to take “the right action,” but what determines which action is right? Prices coordinate effort toward objectives someone has already identified as worth pursuing. AI inherits this limitation. AI can make execution cheaper once a problem is well framed. It can’t tell you which problems are worth framing.

Simon named the first constraint: bounded rationality. Given a clear objective, you can’t compute all paths to it. We built centuries of infrastructure around this — universities, laboratories, scientific methods. We have much less infrastructure for deciding which questions to ask. I call this constraint bounded articulation: given a world full of conditions, we cannot frame all of them into actionable problems. Ruskin put it a century and a half ago: thousands can think for one who can see.

Bounded rationality limits our ability to find better paths to a given end. Bounded articulation limits our ability to turn conditions into problems worth pursuing. The first process can converge on a better answer; the second requires commitments about what matters, rather than a single correct framing to converge on. AI relaxes the first constraint. It cannot make those commitments for us.

And as bounded rationality loosens, evaluative questions proliferate. We can sequence a genome for under a thousand dollars; that exposes questions about genetic privacy and enhancement that no sequencing machine adjudicates. We can build systems that diagnose medical images and draft legal arguments; those capabilities expose what professionals are for once their knowledge advantage dissolves. Better answers don’t settle which questions matter. They reveal how many unsettled values were hiding behind the difficulty of answering.

The displacement scenarios investors model assume a fixed set of problems: as solutions get cheaper, the work required to produce them contracts. That logic holds if the problem set stays fixed. Washington is the template: the I-732 coalition left the evaluative question unresolved, which meant the next technically sound proposal faced the same obstacle. Multiply that across every domain where good answers exist without settled questions to point them at. When questions of direction go unaddressed, what becomes measurable becomes salient, what becomes salient becomes legitimate, what becomes legitimate becomes binding. The visible displaces the valuable.

A firm’s value depends on the problem it chooses and how well it executes. As firms gain access to the same foundation models and tools, AI narrows differences in execution. The choice of problem matters more. Two firms with identical capabilities pointed at different problems produce wildly different outcomes, and the divergence was determined before any computation began.

The same logic scales. “We should care about aging” is a sentiment. “Dignity in aging means someone over seventy can walk to a grocery store, and we’re willing to pay for the infrastructure” is a problem — one that generates answerable questions about measurement, intervention design, workforce development, scaling. Every problem framed is a new source of work. The path out of those displacement scenarios begins with framing more problems worth working on.

I’m not confident we have the institutional infrastructure for this work at the scale our execution capacity now operates. The machines can help us think through a problem. What they cannot supply is the commitment that makes it a problem: something must matter to the one who looks. A condition becomes a problem when someone decides it is unacceptable; computation alone does not make that decision. The economy is producing optimizers at unprecedented scale. What it lacks is seers.