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.
Artificial intelligence (AI) is changing work quickly, but HR should ask one question before automating a task: What capability did this task used to build?
That question matters because some work that looks inefficient on a process map is also where employees learn judgment, context, and quality. Early drafts, basic analysis, research summaries, first customer responses, and routine problem solving may be slow. They may need manager review. They may not look strategic. But for many employees, especially early career employees, this is how work becomes understandable.
AI can now produce many of these first versions in seconds. That can create real value. It can reduce administrative pressure, help employees move faster, and free time for higher value work. But if HR and business leaders remove too much early practice without replacing it with better development, they may solve one productivity problem while creating a future talent problem.
AI Is Reshaping Jobs, Not Only Replacing Them
SHRM’s The State of AI in HR 2026 report found that 39% of organizations have adopted AI in their HR functions, with another 7% intending to launch AI this year. SHRM’s related CHRO analysis also notes that AI is more likely to reshape roles than eliminate them, with 39% of organizations reporting shifts in job responsibilities, 24% reporting new roles, and 57% seeing increased upskilling and reskilling opportunities.
That is the right frame for HR. The main question is not only which jobs disappear. It is how jobs change, what learning disappears inside that change, and what must be rebuilt intentionally.
A junior employee who once prepared the first version of a memo learned more than formatting. They learned how to structure a point, where information was missing, and how a manager evaluated quality. An employee who drafted a first response to a customer learned tone, risk, and context. Someone who handled routine analysis learned where numbers can mislead.
If AI takes over the first version, employees may move sooner into reviewing and approving outputs. That requires judgment, but judgment is usually built through repeated practice, feedback,and mistakes.
HR Should Separate Busywork from Developmental Work
Not every task should be protected. Some work is simply repetitive and should be redesigned. The challenge for HR is to distinguish between work that is only operational and work that also develops capability.
Before automating a task, HR teams can ask:
- What did employees learn by doing this work manually?
- Did the task help build judgment, communication, problem solving, or stakeholder awareness?
- If AI now performs the task, where will employees practice those capabilities?
- How will managers observe whether people are improving?
What new feedback loops are needed?
These questions fit directly into the broader workforce planning challenge SHRM describes in its coverage of the new era of workforce planning, where HR leaders must decide which roles to reskill, redesign or replace as AI changes work.
Replace Lost Practice with Better Practice
The answer is not to preserve outdated work for sentimental reasons. The answer is to design better practice.
HR can help managers build deliberate learning moments into AI-enabled workflows. Employees can be asked to compare AI outputs, explain what they changed, identify missing context, or defend a final recommendation. Teams can use simulations, peer review, manager coaching and decision reviews to replace some of the learning that routine work once provided naturally.
This is where HR has a strategic role. AI adoption should not only measure speed, cost savings, or tool usage. It should also measure whether employees are becoming more capable.
If organizations automate early work without redesigning development, they may find themselves with faster workflows but weaker talent pipelines. The real goal should be better work and stronger people at the same time.
Before HR automates another task, it should ask what that task was teaching. If the lesson still matters, the organization needs a new way to teach it.
Dmitry Zaytsev is the founder of Dandelion Civilization and writes on AI, workforce transformation, HR technology and the future of human capability at work.
Was this resource helpful?
Validate your HR expertise
Earning your SHRM-CP credential makes you a recognized expert and leader in the HR field.
Related Content
Transform HR with mobile technology to streamline workflows, enhance communication, and boost employee engagement across distributed teams.
A new study found that talent development practitioners believe performance-based assessments are more effective at measuring employee learning outcomes than knowledge tests.
As AI makes deeper inroads into organizations, employee concerns about the technology continue to grow. Here's how HR leaders can address the fear of AI among their own staff, as well as in the wider organization.