Teachers in the Age of AI

Luis von Ahn, CEO of the language learning app Duolingo, recently offered the view that AI should replace teachers. He asserted that when AI, rather than humans, is teaching our children, “you'll just see a lot better learning outcomes”—but that schools will not “go away because you still need child care.” I disagree. I think he’s wrong about AI replacing teachers, about how kids learn, and about what makes them want to.

As Rebecca Winthrop of the Brookings Institution said in her recent podcast interview with Ezra Klein, “People always think of education as a transactional transmission of knowledge, which is one important piece of it. But it is actually so much more than that… Kids learn in relationship with other humans. We’ve evolved to do that.”

Teachers observe students as they learn—who is excited and engaged, who is bored and tuning out. They understand how powerful learning can be when students engage with one another in solving a problem, working on a project, or participating in a whole-class discussion. Ultimately, the teacher’s presence and genuine care for students are paramount. I know this as an education technology leader and as a father: When students feel that their teacher truly understands and supports them as individuals, they engage more deeply, embrace productive struggle, and strive to do their best work. Teaching and learning become a shared journey—one that is fundamentally social and truly centered on human connection.

But education does need to change. Teachers need radically stronger supports to do a dramatically harder job, and technology can be an important part of that.

Student learning outcomes, far from bouncing back post-pandemic, remain at the lowest levels in decades. Teachers, although frustrated and exhausted, have expressed optimism in our surveys about the potential of technology to help them. While technology has enabled a 1:1 ratio of devices to students in many places, the reality remains that one teacher is often responsible for 30 or more students. It is simply not feasible for a teacher to respond optimally to the unique learning needs of every student in a large classroom. But Mr. von Ahn's vision—of students spending entire class periods in front of screens with AI tutors instead of human teachers—is dystopian.

But is there a role for AI in education beyond personalized knowledge delivery? The answer lies in complementing, not replacing, the human relationships at the heart of effective teaching.

Currently, teachers rely on three main types of instructional routines in classroom settings, but these routines are often not coordinated in a way that effectively addresses the diverse needs of every student. As a result, many learners may not receive the individualized support or targeted instruction they require to achieve their potential.

1. All together: Whole-class instruction where a teacher delivers a lesson and engages students in activities such as a discussion.
2. Some of us together: Students break up into groups. One group will work with the teacher to practice what we learned when the whole class was together. The other two groups of kids work on projects or problems together that the teacher has assigned to them.
3. On my own: Students work independently to read a book or on adaptive software with an AI Tutor or a writing assignment.

The signature element of the first two instructional routines is humans learning in relationship with other humans. The challenge for the teacher is orchestration of these three instructional routines in a way that is responsive to the needs of every student. Generative AI can support teachers to orchestrate both analog and digital teaching and learning activities optimally for each student. AI can also capture data from learning activities and synthesize that data for the teacher so they can adapt and fine-tune their whole class, group, and independent learning plans to the needs of each student. This creates a learning environment where technology extends the reach of a teacher to meet each student where they are while enhancing human relationships rather than diminishing them.

Perhaps most profoundly, AI can help teachers more easily manage diverse learning needs. And when AI handles administrative and analytical tasks, teachers have more time and energy to invest in meaningful interactions with students. At HMH, we envision a future where technology and human connection work hand in hand, leveraging the strengths of both to create richer, more effective educational experiences for every student.

In short, AI should not replace teachers. But it can help them make a hard job more manageable.

Teaching and learning are fundamentally social processes, rooted in human connection and interaction. At the heart of effective education is a teacher whose role extends far beyond simply delivering content. Teachers are facilitators, mentors, and creators of engaging and effective learning experiences that foster meaningful relationships.

Be excited about the possibilities for AI to support learning. Beware of anyone who tries to sell you on a vision of AI replacing teachers.

***

For more articles on trends in education from experts, subscribe to HMH's LinkedIn newsletter, Policy in Motion.