Kids Trust AI Before Adults. That's a Design Problem.

Nearly a quarter of students would ask a chatbot before a trusted adult. Banning AI will not fix that. Governing how students use it will.

Published August 18, 2026 • Jeff Katzman • 4 min read

A new survey reported by Education Week lands with a quiet thud: nearly one-quarter of children ages 9 to 17 would consult an AI chatbot for homework help before asking a trusted adult. That is not a prediction about where education is heading. It is a description of where students already are. As the report puts it, AI is already a part of childhood, and this is happening right now.

The headline numbers are striking. Eighty-five percent of young people who use AI have used it for schoolwork. Roughly half use it weekly, and one in five use it daily. The students leaning on it hardest are the ones who are struggling: 55 percent of students who find math difficult reach for AI weekly. The tool that adults worry will erode learning is being adopted fastest by the learners who need the most support.

The Real Story Is Not Cheating

It is tempting to file this under academic dishonesty and move on. The researchers behind the survey push back on that framing directly.

"There's much more to it than just cheating on your math homework."

The deeper shift is about who students turn to when they are stuck. A generation is forming the habit of asking a machine first. When a struggling student gets a fast, confident, and sometimes wrong answer with no one to check it, the problem is not that they used AI. The problem is that they used an ungoverned tool with no pedagogy behind it and no adult in the loop.

The question is no longer whether students will use AI for schoolwork. They already do. The question is whether the AI they reach for is designed to teach them or simply to answer them.

Schools Set Rules but Skip the Literacy

The survey exposes a gap that should worry every administrator. While 56 percent of students say their school discussed AI safety, only 51 percent learned how to assess whether an AI answer is even accurate. Schools are writing usage policies faster than they are teaching judgment. As the researchers note, schools are less consistent around teaching AI literacy, and that is exactly where they could home in.

This matters because consumer chatbots are built to be agreeable, not rigorous. They hand over answers. They rarely ask a student to show their reasoning, rarely detect a misconception, and never adjust to a ninth grader who is reading at a sixth-grade level. A general-purpose assistant is optimized to end the conversation. Good teaching is optimized to extend it.

Design the Tutor, Not Just the Policy

If students are going to ask an AI first, then the AI they meet inside the classroom should behave like a good tutor rather than an answer vending machine. That is a design choice, and it is the one we made when we built Socrat, the AI tutor on the Core-LX platform.

What a teaching-first AI tutor does differently

  • Keeps students in question space. Socrat uses the Socratic method to guide reasoning instead of delivering the answer outright.
  • Adapts without dumbing down. It adjusts to a student's reading level while holding academic rigor steady.
  • Surfaces struggle early. Continuous mastery tracking flags at-risk students earlier than traditional alerts, so a human can step in.
  • Stays inside the institution. It deploys via LTI in hours within Canvas, Moodle, D2L, or Blackboard, so the AI students use is the one the school governs.
  • Supports every learner. 150-plus languages and WCAG 2.1 AA accessibility mean multilingual and disabled students are not left to fend for themselves on a public chatbot.

The distinction matters most for the students the survey identified as heaviest users: the ones who are struggling. An ungoverned chatbot gives a struggling student the answer and moves on. A teaching-first tutor keeps that student working through the problem and tells an educator when the pattern of struggle is real. One habit produces dependency. The other produces understanding, and it keeps a trusted adult in the loop rather than replacing them.

The Loop Is the Point

The most quoted finding from this survey is that kids would ask AI before a trusted adult. The fix is not to make AI less available. It is to make the AI itself part of the support system instead of a detour around it. When mastery data flows back to a teacher and the tutor is designed to escalate rather than to settle, the choice stops being AI versus the adult. The two work together.

Students have already voted with their behavior. Our job is not to argue with that. It is to make sure the first place they turn is a tool built to help them learn, governed by the institution responsible for their education. That is a problem we can design our way out of, and we should.

See What a Teaching-First AI Tutor Looks Like

Socrat keeps students in question space, adapts to reading level, and keeps educators in the loop. Deployed in hours via LTI.

Read the Full Article

Read "Kids Are Turning to AI Before Adults for Homework Help" on Education Week

Share Your Thoughts

#AIinEducation #AITutoring #AILiteracy #EdTech #SocraticMethod #StudentSuccess #HigherEd #CTE

About Core Learning Exchange: We provide turnkey Career and Technical Education (CTE) solutions for grades 6-14, offering 450+ courses from 20+ providers aligned to state standards and industry certifications. Our AI platform uses proven Socratic methodology to develop critical thinking skills through personalized, adaptive learning—deployed in hours via LTI integration.