Write an article about
Key points:
Last fall, during a professional development session I was running with a group of teachers in São Paulo, a fifth-grade teacher raised her hand and asked a question I have since heard in every country I work in: “I want to use AI to plan better lessons. But how do I do that without just putting kids in front of another screen?”
She already knew what the research said. She saw what too much screen time did to her students’ attention spans, to their handwriting, to their ability to sit with a problem that did not have a loading bar. She was not asking me to justify AI. She was asking me whether AI could work in a way that left her classroom alone.
That question is one the current debate is not answering.
The critique coming from researchers like neuroscientist Jared Cooney Horvath deserves to be taken seriously. The footprint of educational technology in K-12 classrooms has grown faster than our evidence base for it. Devices arrived before pedagogy did, and we handed students tools that were engineered to hold adult attention. Cooney Horvath and others who share his concerns are pointing at something real.
But here is what troubles me about where the conversation goes from there: The critics and the edtech industry have ended up agreeing on the same premise. Both assume that AI in education means AI in front of students. The critics say that is dangerous, and they are right. The industry says it is inevitable, and keeps building that way. Neither side has stopped to ask whether the premise is the problem.
The blind spot is this: AI does not have to face students. It can face the teacher.
I have spent 15 years training teachers–mostly in Brazil, but also in North America. The number one thing teachers ask for is not a smarter student-facing app. It is time and structure. Help me build the lesson. Help me find the right activity for a class that is half kinesthetic learners and half kids who shut down the moment they feel embarrassed. Help me design a discussion that does not fall apart in the first five minutes.
That is a prep problem, not a delivery problem. And it is exactly the kind of problem AI is well-suited to solve.
Here is what this looks like in practice. A teacher sits down on a Sunday afternoon with a learning objective and a rough sense of her students. She describes her class to an AI tool: the range of levels, the topics that have gone flat before, the kid in the back row who will either derail everything or be the best participant depending on how the opening is framed. The AI helps her build a lesson structure, suggest discussion questions, draft a short formative check. She edits, pushes back, refines. She arrives Monday with a plan that is tighter and more responsive than anything she had time to build alone.
Monday morning, the AI is nowhere in sight. No student accounts. No dashboards tracking engagement metrics on nine-year-olds. No devices open on desks. There is a teacher, a lesson, and 30 kids who are about to have a discussion about something that matters.
This model solves three things at once.
First, it returns time to teachers. Prep is where teachers bleed hours they do not have. A well-structured AI tool for lesson design handles the scaffolding so the teacher can spend that hour on the work only she can do.
Second, it improves lesson quality in the places where teachers most want support. The weakest point in a lesson is usually not delivery. It is structure. The sequencing of activities, the transition between independent and group work, the moment a discussion needs a pivot. These are designable in advance, and AI is quite good at helping design them.
Third, it resolves the false choice that has paralyzed so many principals and district leaders. You do not have to choose between the anti-screen camp and the full-AI camp. You can use AI to make teachers more effective and keep the classroom fully human. Both at the same time.
I want to be honest about what this model asks of teachers, because it asks more, not less. Teacher-facing AI does not simplify teaching. It deepens the prep work. It asks the teacher to think clearly about her learning objectives before she gets to the tool. It requires her to evaluate what the AI produces and push back when it is generic. That is not a shortcut. It is craft with a better starting point.
The point is not to replace the teacher. The point is to give her more of what she actually needs: time and structure, so she can be fully present with students.
To edtech builders: The most useful thing you can build right now is something teachers open at 9 p.m. and students never see. Design for the person who carries the lesson, not the person who receives it.
To principals and district leaders: The next time a vendor puts a device in a student’s hand during a demo, ask what happens if you take it away. If the answer is “nothing, because the teacher still has everything she needs,” that is the product worth your budget.
The screen-time debate has been asking the right question in the wrong direction. The screen we should be thinking about is the teacher’s.
Adriana Perusin, IASEA & Flip Education
Adriana Perusin is a Canadian-Brazilian educator with over 20 years of experience in education and over 15 years training more than 1,000 teachers in active learning and social-emotional skills. She founded IASEA in Brazil for teacher professional development and is co-founder of Flip Education, which builds AI co-teacher tools designed to be used by teachers, not students.
Latest posts by eSchool Media Contributors (see all)
.Organize the content with appropriate headings and subheadings ( h2, h3, h4, h5, h6). Include conclusion section and FAQs section with Proper questions and answers at the end. do not include the title. it must return only article i dont want any extra information or introductory text with article e.g: ” Here is rewritten article:” or “Here is the rewritten content:”
