Before HR Automates a Task With AI, Ask What It Used to Teach - SHRM

TEXT START: Editor's note: Posts published on From the Workplace are written by outside contributors and do not necessarily reflect the views or opinions of SHRM.

The Dissection

This article identifies a real secondary wound: AI removes the routine tasks through which junior workers acquired judgment. It then converts a structural labor-market rupture into an HR design problem—replace lost practice with coaching, simulations, peer review, and decision exercises.

The piece makes AI adoption administratively respectable while leaving the employment system untouched.

The Core Fallacy

It treats lost practice as the central crisis. Under the Discontinuity Thesis, the central crisis is the destruction of economically necessary human labor itself.

Replacing routine work with “better practice” may preserve capability-building for a smaller elite. It does not restore the mass employment → wage → consumption circuit. If AI produces drafts, analyses, responses, and recommendations faster and cheaper, competitive organizations will reduce the number of humans receiving those repetitions—not merely redesign them.

The article assumes a mostly human workforce remains and merely needs better workflows. Under P1, P2, and P3, that assumption is the corpse being cosmetically maintained. Capability is not bargaining power. A well-trained surplus worker is still surplus.

Hidden Assumptions

• Employers will preserve entry-level roles after AI makes their training tasks unnecessary.
• Human judgment will remain scarce enough to support mass employment.
• Managers will provide coaching rather than automate review and supervision too.
• Simulations can reproduce the volume, stakes, context, and feedback of real work.
• “Higher-value work” will expand enough to absorb those displaced from lower-value work.
• Reskilling is a solution rather than a delay mechanism.
• HR can override the competitive pressure toward lower labor costs and higher throughput.

These are institutional hopes, not demonstrated structural protections.

Social Function

Primary classification: transition management and ideological anesthetic, with a substantial partial truth.

The partial truth is that AI can erase apprenticeship embedded in routine work, producing a thinner and more brittle talent pipeline. The anesthetic is the implication that intentional learning design solves the larger problem. It does not. It helps selected workers remain useful while obscuring that the system is steadily reducing the number of workers it needs.

The article also provides elite self-exoneration. Leaders can claim to protect human capability while pursuing automation for speed and cost, then assign HR the task of inventing substitute practice after the economic reason for employing people has been removed.

The Verdict

This is a perceptive hospice memo for entry-level work. It identifies the apprenticeship tissue AI will destroy, then mistakes replacement training for preservation of the patient.

Better simulations may produce sharper servitors and a smaller layer of reviewers, but they cannot reverse productive participation collapse. The likely result is faster production, narrower human bottlenecks, and a growing population trained for roles the system no longer needs.