TEACHER VOICE: The future of personalized learning cannot be built on technology alone

By the time many of my high schoolers enter my algebra classroom, they already know what their school thinks they are bad at. They carry labels before they ever open a notebook: “remedial,” “unmotivated,” “low-performing.”

Too often, personalized learning begins with those labels. We start with a low score or what we believe students cannot do — a missing skill or an academic gap — and then build instruction around fixing it.

I think we should start somewhere else: with what students care about, what they want to become and what already makes sense to them.

The starting point matters because personalized learning is supposed to do more than respond to academic gaps. Personalized learning is supposed to tailor instruction and learning experiences to support a student’s individual strengths, needs, interests, goals and pace.

But whether it fulfills that promise depends on the information we choose to prioritize. That prioritization matters even more as artificial intelligence becomes a bigger part of personalized learning.

AI gives teachers like me new ways to adapt instruction quickly. It can create practice problems, adjust reading levels, provide immediate feedback and generate materials based on student needs. But technology can only personalize the information we choose to give it.

That is why I believe a better approach to personalized learning with AI is giving it information to build learning from the student’s interests and goals. Otherwise, we risk using a new technology to make an old approach more entrenched instead of making learning more meaningful.

I saw what the difference could look like with a student I will call Kyle, who recently asked me, “If my phone can solve this in two seconds, why do I have to learn it?”

It was not a disrespectful question. It was an honest one. In an age of instant answers, more students are likely to wonder the same thing.

I could have responded with the usual answers: Because it is on the test, because you need the credits, because you will need this later.

Instead, I asked him what he cared about. Kyle started talking about cars, music and wanting to make money. That conversation gave us a way into the math.

Once I understood what mattered to him, I could use AI-supported personalized learning differently. I used it to generate examples connected to his interests, like comparing car loans and monthly payments and the long-term cost of financial decisions.

Through his interests, I was also able to give him work that allowed me to learn more about his ability than a worksheet or test score could have shown me. I saw how he reasoned when the context mattered to him. I saw where he got stuck, and I saw what made him willing to keep going. His confidence shifted because the math was no longer something being done to him. It became something connected to his future.

Generative AI can already support personalized learning in all the ways mentioned above. But if we are not careful, AI will sort students according to what they cannot do yet and call that innovation.

That is why I worry when personalized learning is defined only by academic level. Of course, I need to know where students are struggling. A math teacher like me cannot ignore gaps in understanding. But a student who struggles with equations may still be a sharp thinker when the math is connected to business, family responsibilities or sports.

In my classroom, I have learned that interest is not a distraction from rigor. It is often the entry point into it. AI helped me create relevant practice more quickly, but it did not create the reason for learning. That came from Kyle’s own goals and from the conversations we had in the classroom.

Academic data alone cannot tell me who a student is, how they think, what they value or what might make them willing to wrestle with something difficult. Starting with Kyle’s interests did not lower the mathematical expectations. It gave me a better way to help him reach them — with AI’s assistance.

That is the role that I think AI should play in personalized learning.

The starting point should be this: What do my students care about? What are they trying to become? What already makes sense to them?

Then AI can help me personalize the work and find ways to make it matter to them. In the process, my relationship with my students will deepen.

For students like Kyle, starting with who they are rather than what they are missing is what makes deeper learning possible. Knowing what matters to my students changes how I understand what they are capable of.

Al Rabanera teaches math at La Vista High School in Fullerton, California. He is a 2025-27 Teach Plus Leading Edge Educator Fellow.

Contact the opinion editor at opinion@hechingerreport.org.

This story about personalized learning was produced by The Hechinger Report, a nonprofit, independent news organization focused on inequality and innovation in education. Sign up for Hechinger’s weekly newsletter.