The Hidden Cost of AI in Education

April 18, 2026

“If the only tool you have is a hammer, it is tempting to treat everything as if it were a nail.” — Abraham Maslow

Over the past few weeks, I’ve spent a lot of time thinking about AI. Conversations at ASU+GSV and discussions with peers pushed me to dig a little deeper into what’s actually happening beneath the surface.

The more I looked at it, the more something started to feel familiar.

There’s a pattern in education and in technology more broadly. A new capability shows up with the promise of simplifying things, and in many ways it does. But it also changes expectations. And those expectations tend to move faster than systems, policy, or people are ready for.

That’s the part I think we need to spend more time on.

Right now, a lot of organizations are approaching AI like it’s the answer to everything. If there’s a problem, AI becomes the default solution. The hammer.

I understand why. The productivity gains are real. The ability to analyze data, generate content, and streamline workflows is unlike anything we’ve seen before.

But there are costs that don’t show up on a dashboard.

What I’ve seen inside organizations is a shift in the nature of work. AI takes away many of the basic tasks, which sounds like a win. But what replaces those tasks isn’t less work, it’s harder work. More decisions. More data to interpret. More nuance in every interaction.

And when everyone is operating at that level at the same time, something else starts to happen.

Expectations rise.

And to be clear, that’s not inherently a bad thing. Higher expectations can drive better outcomes. But we also have to be honest about the ramifications if those expectations outpace capacity.

The baseline changes from “getting the work done” to “getting the work done faster, deeper, and at a higher level.” On paper, that looks like progress. In practice, it can feel like constant pressure.

I’ve had conversations with people who are starting to quietly ask a different question. Not how AI helps them, but whether it replaces them. At the same time, leadership teams are trying to balance real pressures around efficiency, growth, and margin. I’ve sat in those seats, owning P&L and managing EBITDA. That pressure is real.

But the best organizations won’t just chase productivity. They’ll recognize the tradeoffs that come with it.

You’re already starting to see this play out. Companies like Adobe and others are beginning to put guardrails in place, not to slow AI down, but to ensure there’s still human accountability, balance, and intention behind how it’s used.

This dynamic is already playing out in the private sector. Education won’t be far behind.

School systems have historically lagged slightly behind when it comes to data usage, but not because of a lack of talent. There are brilliant educators at every level. More often than not, it’s policy at the state and local level that slows the ability to move, creating constraints that don’t exist in the private sector.

That matters, because it means the challenge isn’t capability. It’s alignment.

And it also creates a window. A chance to learn from what’s happening elsewhere before fully scaling it inside schools.

Because when this shows up in education, it hits differently.

Teachers already operate in one of the most demanding environments there is. There’s a significant body of research on the number of decisions a teacher makes in a single day. AI has the potential to help with that. It can reduce cognitive load, streamline planning, and surface insights that were previously hard to access.

But if it’s not implemented thoughtfully, it can just as easily increase expectations faster than capacity.

And in a system already dealing with burnout and teacher shortages, that’s not a small risk. That’s a structural one.

At the same time, there is real upside here. Access to high-quality learning is becoming more democratized. The ability to personalize at scale is no longer theoretical. If done right, this could be the transformation education has been working toward for decades.

But that outcome isn’t guaranteed.

It depends on how intentional we are right now.

Districts that get this right will take a holistic view. They won’t treat AI as a tool to deploy, but as a system to align. They’ll stay grounded in student outcomes and make decisions through that lens.

More importantly, they’ll protect their most valuable resource: teachers.

If AI is working the way it should, it gives time back. It creates space. It allows educators to focus more on students and less on everything else that pulls them away from that work.

That requires discipline.

This can’t be a ready, fire, aim moment. It has to be ready, aim, fire. Define the outcome first, then apply the technology with purpose. AI is quickly becoming a commodity. The advantage won’t come from using it. It will come from how it’s used.

The organizations that win will be the ones that stay focused on outcomes, not tools. The ones that create environments where people can do their best work, not just more work.

Because AI doesn’t just create leverage.

It creates pressure.

And we’re only starting to understand the cost of that.