The HR Function Is Breaking Under the Weight of Work It Was Never Staffed to Do

The most revealing sentence in any board presentation about AI transformation isn't about technology. It's about who is supposed to implement it. At most organizations, the answer is the same human resources (HR) team that is already overextended, under-resourced, and expected to simultaneously manage hybrid work tension, compliance changes, and leadership pipelines — without proportional investment.

According to Lattice, as cited by HR Dive (2025-07-14), HR teams report that their job demands have grown substantially, but the resources allocated to meet those demands have not kept pace. That single finding — HR job scope has expanded while resourcing has not — should stop every chief executive officer (CEO) who has handed a "people transformation" mandate to their chief human resources officer (CHRO) without a corresponding budget conversation.

Why are HR workloads rising faster than HR headcount?

The answer isn't that HR is inefficient. It's that the job has been fundamentally redefined by forces that arrived simultaneously: generative AI adoption, the normalization of remote and hybrid work, and a leadership pipeline that most organizations cannot honestly say is healthy. Each of these forces, alone, would require serious organizational investment. Together, they have created a category of work that didn't exist five years ago — and for which no legacy HR operating model was designed.

Consider what generative AI alone is doing to organizational stability. MIT Sloan Management Review (2025-07-14) describes a vice president (VP) of product who arrives on a Monday morning to find that the AI model her team spent six weeks integrating into a customer workflow has already been superseded by a cheaper, faster alternative. That cycle of obsolescence — which MIT Sloan calls "AI disruption that never ends" — doesn't just affect product teams. It lands directly in HR's lap: retraining curricula go stale, job architectures built around specific tools become obsolete within quarters, and the psychological contract with workers who were promised "AI-augmented" roles gets renegotiated without warning.

What does permanent AI churn actually cost the HR function?

Generative AI is not a one-time implementation project. It is a continuous change-management obligation. MIT Sloan Management Review (2025-07-14) notes that approximately 2.4 billion people worldwide use generative AI platforms each month as of 2025 — a scale that makes AI literacy not a training elective but a baseline organizational competency requirement. Someone has to own that. In most organizations, that someone is HR, operating without new headcount, new budget authorization, or a clear mandate from the executive team.

Generative AI (gen AI) is software that produces text, code, images, or other content by predicting statistically likely outputs from patterns in training data. It is not a database lookup or a rules engine — it produces novel outputs that require human judgment to evaluate. That distinction matters for HR because it means workers cannot simply "learn" gen AI once; the tools change faster than annual training cycles, demanding a fundamentally different approach to capability building.

The compounding problem is that remote and hybrid work — still unresolved for most organizations — is independently driving demand for the same HR capacity. Research covered by HR Executive (2025-07-14) finds that both AI adoption and remote work are intensifying demand for interpersonal skills, the exact skills that are hardest to develop asynchronously and hardest to assess without behavioral data. HR teams are being asked to build coaching cultures and leadership pipelines at the same moment their capacity to do so is being consumed by compliance, tool-stack churn, and organizational anxiety.

The strategic implication here is not that HR needs better productivity tools. It is that most executive teams have made a category error: they have treated HR as an implementation arm for transformation initiatives rather than as a function that itself requires transformation investment. When the CHRO's team is the one expected to manage AI change while simultaneously being asked to absorb the change, you have not built an AI strategy — you have built a pressure cooker.

The decision that actually matters isn't which AI vendor to choose or which return-to-office policy to enforce. It's whether your organization is willing to resource the function responsible for managing all of it — or whether you are quietly asking HR to do impossible work and planning to be surprised when it fails.