The cycle that began with [YOU] on AI documents the technology trap from the inside: the senior engineer whose architectural judgment is more valuable than ever but whose professional standing has not adjusted to reflect this, because the informal norms governing professional valuation were designed around a production function in which implementation competence and architectural judgment were bundled together. The tool has unbundled them. The norms have not.
The cycle’s call for educational institutions to teach questioning over answering, integration over specialization, and judgment over execution is precisely the kind of institutional redesign the technology trap demands. But North’s framework insists on a corollary that the cycle underemphasizes: the call underestimates the difficulty of the change, because it does not adequately account for the structural forces that hold the existing institutions in place. Path dependence is not overcome by good arguments. It is overcome by changes in the incentive structures that make the existing path self-reinforcing—changes that require not just vision but institutional entrepreneurship, political coalition-building, and the patient work of redesigning systems while the systems are running.
North developed the technology trap implicitly across his body of work on path dependence and institutional change, but the explicit framework draws on Paul David’s 1985 paper on the QWERTY keyboard—which demonstrated that the layout had been determined by an early engineering constraint that no longer applied, and that the layout persisted not because it was optimal but because switching costs exceeded benefits for any individual typist. North extended the analysis from technology to institutions at vastly greater scale and with far greater consequences: institutional path dependence operates through increasing returns, growing constituencies, and accumulated investment that makes the framework progressively more resistant to modification.
The AI-specific instantiation of the technology trap connects to the broader framework of the institutional void: the technology trap describes the persistence of old institutions past their useful life, while the institutional void describes the absence of new institutions adequate to the present. Both conditions coexist in the current moment, and together they define the institutional challenge of the AI transition.
Rational persistence of irrational institutions. The institutions of the technology trap were rational when designed. They resist change not through irrationality but through the rational calculus of actors who have invested in the existing framework. The law firm that has built expertise in employment regulation, the school system that has built curriculum around standardized competencies, the professional association that has built gatekeeping around pre-AI skills—each has a rational interest in the persistence of the framework that made those investments valuable.
Three domains of AI-era path dependence. Employment law assumes time as a proxy for productive contribution—a model that the twenty-fold productivity multiplier makes obsolete. Educational systems were designed to produce standardized cognitive workers—precisely the workers AI replicates most easily. Professional licensing certifies competencies that AI tools can replicate at marginal cost, while the competencies that matter most in an AI-augmented practice (the ability to evaluate machine output) are not what licensing examinations test.
The conditions for escape. North’s later work identified the conditions under which institutional change occurs despite path dependence: shifts in relative bargaining power that alter the incentives of actors within the existing framework, creating opportunities for institutional entrepreneurs to propose and implement new arrangements. The AI transition is producing exactly such a shift. The window for influencing the emerging framework is finite. Path dependence will lock in whatever arrangements emerge.