Rebecca Henderson

In the AI Story

Henderson's intellectual formation combined rigorous economic training with a sustained engagement with organizational and institutional questions that pure economics often sidesteps. Her early work on architectural innovation drew on Herbert Simon's bounded rationality, James March's organizational learning, and Thomas Kuhn's paradigm shifts—synthesizing cognitive, organizational, and competitive dynamics into a framework that became one of the most cited in innovation studies. The synthesis was not merely interdisciplinary borrowing. It was a structural insight: that the most consequential innovations operate at the level of relationships rather than components.

The turn toward purpose and sustainability that defined her later career appeared to some colleagues as a departure from the structural rigor of the early work. Henderson has consistently argued that it is an extension of the same insight to a larger canvas. If architectural knowledge determines organizational success or failure at the firm level, it also determines institutional success or failure at the societal level. Capitalism encodes architectural assumptions—about externalities, time horizons, stakeholder obligations—into its structures. When those assumptions become inadequate to the world, the system fails in precisely the way that incumbent firms fail: not from malice or incompetence, but from expertise that has become a liability.

Henderson's work on AI represents the convergence of both strands. The 2018 paper with Cockburn and Stern applied the architectural innovation framework to AI as a general-purpose technology, predicting racing dynamics, concentration of capability, and the need for policies encouraging data-sharing. Her analysis of AI in Reimagining Capitalism extended the purposeful-capitalism thesis into the AI era, arguing that firms deploying AI within extractive architectures will optimize themselves into fragility, while firms deploying AI within stakeholder-accountable architectures will create durable value.

Henderson serves on multiple corporate boards—including Amgen, Idexx Laboratories, and Cummins—providing a vantage on how the architectural frameworks she studies operate in practice. She advises governments and international organizations on innovation policy and climate governance. Her teaching at Harvard Business School has shaped a generation of executives, many of whom occupy positions where architectural decisions about AI deployment are being made. The frameworks she built are not academic exercises. They are operational instruments, tested against the resistance of real organizations operating under real competitive pressure.

Origin

Henderson's 1990 breakthrough emerged from intensive fieldwork on an industry few people had heard of: photolithographic alignment equipment. The choice of industry was strategic. It was technically complex, data-rich, and had undergone multiple architectural transitions whose outcomes were observable. The puzzle that drew her was the persistent failure of established firms—not occasional, but systematic, predictable, and inexplicable within existing innovation theory.

The puzzle's resolution required a conceptual innovation as consequential as the empirical findings. By introducing architecture as a dimension orthogonal to component change, Henderson created a framework that explained not just why incumbents failed but why their failure was structurally predictable. The embedded architectural knowledge that made them successful in one generation became the filter that blinded them in the next. The insight was portable across industries and decades, and it has proven disturbingly applicable to the AI transition, where the most expert organizations are the most vulnerable to the architectural shift AI represents.

Key Ideas

Architectural knowledge as embedded filter. Organizations encode assumptions about component relationships into their structures—making architectural shifts invisible to the institutions whose survival depends on perceiving them.

Purpose as competitive architecture. Purpose is not moral sentiment but structural feature—the embedded understanding of how value-creation components relate to each other and to broader systems, determining what signals the firm sends the AI amplifier.

Internalization of externalities as market correction. Markets produce efficient outcomes only when costs and benefits are captured in prices—internalizing externalities is not anti-market but the condition under which markets function as their advocates claim.

Stakeholder accountability as institutional architecture. The redesign of capitalism to hold firms accountable to employees, communities, suppliers, and the environment—not through moral exhortation but through structural mechanisms that make stakeholder impact visible in decision architectures.

The engagement advantage. New architectural knowledge can only be built through direct interaction with the new architecture—observation from within the old architecture filters out the signal, making early engagement structurally advantageous.