Augmentation versus automation is the framing that distinguishes between using AI to replace what a worker does and using it to amplify what a worker can do. Automation substitutes a machine for a human task; augmentation pairs the machine with the human so the human becomes more productive at the tasks they keep. The distinction matters because the two paths can produce very different outcomes for wages, employment, and who captures the value created.
David Autor’s work, especially his 2015 essay “Why Are There Still So Many Jobs?,” supplies the economic backbone for the augmentation view. He shows that technology has historically both substituted for labor in routine tasks and complemented it in non-routine ones, so the net effect on jobs depends on which force dominates and on whether new tasks are created. Daron Acemoglu and Simon Johnson push the argument further in their book “Power and Progress,” contending that whether a technology augments or automates is not predetermined by the technology itself but is a choice shaped by the priorities of those who design and deploy it. An automation-first bias, they warn, can concentrate gains among capital owners while doing little for workers.
For business leaders the framing is directly actionable. The same AI system can be deployed to cut headcount or to make existing staff more capable; the choice affects productivity, morale, and the distribution of returns. The augmentation case is not merely ethical, it is often the more durable source of competitive advantage, because it builds on human judgment rather than trying to eliminate it.