One Teacher’s Story
A social studies teacher friend of mine told me recently that AI has changed her work in ways she did not expect. She was an early adopter, using AI to strengthen feedback and make grading more efficient. At first, it worked, allowing her to generate feedback that was more specific, individualized, and actionable than a rubric or checklist alone typically allowed in the past. But as that became easier to produce, expectations shifted. Detailed feedback was no longer optional. It had quietly become the norm, less because of administrators and more because AI made it possible, and therefore expected by students and parents.
But easier did not mean lighter. She still had to review, vet, and revise everything. She was still accountable for every word.
“It’s better for kids,” she said. “But I’m more exhausted than before.”
That may be the irony of AI in schools. The very tool promoted as a way to save time can end up expanding the work.
What the Research Is Starting to Show
What this teacher described is not an isolated experience. A recent article in Harvard Business Review argues that AI does not reduce work, but intensifies it. Workers complete tasks more quickly, but instead of reclaiming time, they take on more. The result is greater cognitive load, longer work hours, and expanded responsibilities.
AI increases speed, but it does not necessarily make work feel lighter because expectations rise alongside efficiency. Schools are not outside this pattern. If anything, they amplify it.
When Efficiency Expands the Work
The introduction of AI into school environments often leads to a rapid shift in baseline expectations. As efficiency increases, so does the scope of work; lesson plans require deeper differentiation, communication remains constant, and data analysis becomes both more frequent and granular. This evolution redefines the standard for quality, where previously comprehensive work is now viewed as merely basic.
While higher standards can benefit students, a problem emerges when these increased demands are viewed as resource-neutral. Educators are often forced to manage expanded workloads without additional time or guidance. Rather than simply saving time, efficiency recalibrates what is expected of staff, effectively filling any newly created capacity.
This expansion of workload is evident across several key areas:
- Instructional Planning and Grading: The ease of AI generation leads to greater demands for personalized feedback and highly customized lesson materials.
- Stakeholder Communication: As parents utilize AI to craft increasingly formal and legalistic messages, educators face a growing burden of processing and interpreting these communications. Even with AI assistance in drafting replies, the cognitive effort required to process the original messages remains high.
- Administrative Documentation: The time saved through automation is frequently redirected toward increased documentation and administrative requirements.
The Loss of Natural Limits
Teaching has never been a light profession. Even before AI, educators managed heavy workloads that required constant planning, decision-making, and responsiveness. What existed, however, were natural limits within the work–time between classes, the pace of planning, and the effort required to produce materials created built-in boundaries. Those pauses did not reduce the workload, but helped contain it.
As the same teacher explained, she used to have small pauses between classes. Now AI makes it easier to fill those moments. She might answer an email between periods, draft something during lunch, or use a few free minutes to produce one more item before the next part of the day begins.
What has changed is not the behavior, but the unspoken expectation. Because AI makes more possible in less time, those pauses no longer feel like pauses, but become opportunities to produce more.
This is where the shift becomes most visible. Work moves faster and becomes more dense, with fewer natural stopping points, squeezing out recovery time and reflective space. Cognitive load increases and decision fatigue builds, making burnout more likely.
We often focus on what AI makes possible, but spend less time examining what it requires from the people using it. This is not just a question of adoption, but one of sustainability.
What Leaders Must Do
If AI is going to remain part of school systems, then leadership has to define and shape how it is used. It is not enough to benefit from faster work.
- Clarify when AI should be used. Not every task should be automated. Leaders must distinguish between efficiency and professional judgment.
- Set limits on what speed should trigger. Faster output should not automatically lead to more output or higher expectations.
- Protect time for thinking. Schools need space for reflection and decision-making, not just continuous production.
- Match rising expectations with real support. Training, time, clarity, and resources must align with what is being asked of staff.
- Eliminate or streamline existing requirements. If AI enables more detailed feedback or expanded expectations, something else has to give. Leaders should identify tasks that can be reduced or removed, such as redundant documentation, excessive grading requirements, or rethinking how rubrics and checklists are used. Without subtraction, every gain becomes an addition.
- Keep the human at the center. AI can support practice, but educators remain responsible for accuracy, judgment, and care.
Leaders cannot raise the bar and then leave people to clear it alone.
Closing Thought
AI does not remove work, it removes friction, and with it, some of the limits that once kept work contained, which can lead to better outcomes. Without thoughtful leadership, it can also lead to faster burnout.