Editor’s note: This essay is an entry in Fordham’s 2026 Wonkathon, which asked contributors to answer this question: “What rules should govern the uses of digital technology in schools—and who should set them?” Learn more.
We are asking the wrong question about technology in schools.
Debates over screen time, AI, and devices obscure a more fundamental question: What kind of human beings are we trying to help young people become?
Education develops more than knowledge. At its best, it cultivates reasoning, judgment, understanding of ourselves and others, and meaningful contribution.
Artificial intelligence presents an extraordinary opportunity to advance an even deeper purpose of education: the development of human consciousness. By consciousness, I mean not only awareness of our thoughts, but the capacity to direct attention, examine our assumptions, and become more attentive to the people and world around us.
AI could cultivate these capacities or undermine them. The difference depends on the rules and purposes we choose.
The risk of making learning too easy
Consider two students encountering a difficult mathematical problem. One asks AI for the answer, receives an explanation, and moves on. The other wrestles with the problem, makes mistakes, notices her frustration, tries another approach, and eventually arrives at a solution.
Both may produce the correct answer. But the second student has practiced far more than mathematics: attention, persistence, intellectual humility, and the ability to remain engaged in uncertainty.
AI can remove cognitive effort from almost any task. Sometimes that is exactly what we want. A student with a reading disability may benefit enormously from accessible text. But when technology routinely removes the struggle through which understanding develops, students may generate sophisticated answers without becoming capable thinkers.
The purpose of educational technology should not be to eliminate difficulty. It should be to make productive difficulty more accessible.
From artificial intelligence to human consciousness
This is the possibility I find most exciting: using artificial intelligence to cultivate the distinctly human capacities that intelligence alone cannot guarantee.
In high school, I read Daniel Quinn’s Ishmael, and something shifted. I suddenly became aware of assumptions and inherited stories shaping how I understood the world. I remember how intensely I resisted ideas that contradicted them, even while realizing many of my convictions were not truths I had experienced or tested myself.
AI might have helped me trace those assumptions, test competing evidence, and examine why certain ideas provoked such resistance. Or it might have handed me a neat summary of the book and ready answers, letting me bypass the discomfort that made the experience transformative.
Imagine a student using AI to develop an argument. Rather than write it for him, AI challenges his conclusions, offers contradictory evidence, and asks him to defend his reasoning. In mathematics, it might prompt another approach rather than supply the solution. In a classroom disagreement, it might help students separate what they observed from what they assumed.
These experiences deepen effort and help young people notice not only what they think, but how their minds work. That awareness might help them move beyond self-concern toward curiosity, connection, and contribution.
An educational AI tool should be evaluated not only by what it enables a student to accomplish while using it, but by what it enables the student to experience, understand, and do when it is no longer present.
Five rules for governing technology in schools
1. Protect attention. Attention is the foundation of consciousness and learning. Schools should block social media feeds, advertising, games, and unnecessary notifications on instructional devices, and keep personal phones away during the school day. In elementary grades, printed books, handwriting, discussion, and physical exploration should be the default. Rather than assigning every young child a device, schools should use shared devices for a clearly defined purpose. In later grades, teachers should close devices when they are not needed. The goal is not simply less screen time, but greater control over attention.
2. Preserve intellectual independence. Students must regularly demonstrate what they can understand, create, and explain without AI. Schools should distinguish work AI may support from work students must first attempt independently. Require first drafts, worked solutions, oral explanations, and some assessments completed without digital assistance. AI can critique an argument or provide targeted practice, but should not produce the thinking students are meant to learn.
3. Evaluate technology by the human capacities it develops. Before buying or renewing a tool, districts should identify the instructional need, compare it with a credible alternative, and specify what learning will improve. Pilot tools, examine student work and achievement, ask teachers about time and distraction, and test whether skills transfer when the device is closed. Publish an annual inventory of costs, usage, and evidence; discontinue tools that fail to justify their instructional time. Engagement is not the same as learning.
4. Keep human relationships at the center. Belonging, empathy, and moral judgment develop through relationships with real people. AI may help students prepare for conversations, not replace teachers, counselors, or peers. Schools should favor face-to-face dialogue and collaborative work whenever those are the learning goals. Consequential decisions about students must remain under human authority, with strong privacy protections. Devices should go home only for a defined learning or accessibility need, with families informed and printed alternatives where practical.
5. Teach students to question AI, including when not to use it. AI literacy must extend beyond prompting. Students should learn to detect errors and bias, challenge persuasive answers, and examine how platforms shape their attention and beliefs. Have them compare AI responses with primary sources, identify what a tool got wrong, and explain why they chose to use it. The ultimate measure of AI literacy is not proficiency with AI. It is the wisdom to use it intentionally.
Who should make the rules?
Governance should follow a simple principle: Establish common protections centrally, while keeping instructional judgment close to students.
States should set minimum standards for privacy, transparency, commercial exploitation, accessibility, and age-appropriate safeguards.
Districts should establish procurement standards, evaluate evidence, train educators, and adopt clear expectations for appropriate AI use. Families and students should have meaningful input.
Teachers and school leaders should retain flexibility to determine which instructional approaches best serve their students, including appropriate accommodations for students with disabilities and English learners, and technology-focused courses.
Students should gradually assume greater responsibility for governing their own technology use. After all, the goal is not permanent adult control. It is developing adults capable of governing themselves.
What it means to become more human
In an era when information and sophisticated answers are increasingly effortless, attention, judgment, courage, compassion, and wisdom will still require cultivation.
If education is primarily about producing correct answers, AI may outperform much of what we call schooling. But if education is about developing conscious, capable, compassionate human beings, the possibilities are extraordinary.
