67% of Students Say Heavy AI Use Is Hurting Their Own Thinking — and They Keep Using It Anyway
Back to blog
Family Safety·OCTOBER 13, 2026

67% of Students Say Heavy AI Use Is Hurting Their Own Thinking — and They Keep Using It Anyway

The Blaick Team 5 min read

Children aren't blind to the trade-off. In a nationally representative RAND survey of 1,214 young people, a clear majority said that leaning on AI for schoolwork harms their critical thinking. The unsettling part is that many of them are doing it anyway.

The number to hold: 67%. That is the share of students who agreed or strongly agreed that heavier AI use for schoolwork harms critical thinking. Even among students using AI in three or more classes, 53% still said so.


Middle schoolers
68% worried
The youngest group surveyed was the most likely to say heavy AI use hurts thinking.
Heavy users
53% still agree
Students using AI in three or more classes say it harms their thinking.
Non-users
78% agree
Students who avoid AI are the most convinced it damages thinking.
Strong worry
27% strongly agree
More than one in four say they feel it strongly.

Source: RAND survey of 1,214 young people, fielded December 2025 and published August 2026. RAND is careful to note this is a self-reported belief, not a measured change in how well anyone actually thinks.


A Worry Is Not a Verdict

It's worth being fair. These numbers describe what students believe, not a brain scan or a test score. Some studies of carefully designed AI tutoring show modest gains. The point isn't that AI makes children less capable. It's that children themselves sense something is off, and that is a signal parents can use.

The tension in the data

Awareness isn't restraint. Knowing the risk doesn't stop a tired kid with a deadline from pasting in the question.

Tools are built to be easy. A chatbot that hands over a finished answer is always more tempting than the effort of working it out.

It looks like learning. A polished answer on the page can hide that nothing was understood along the way.

Your child may already suspect AI is thinking for them. They're waiting for an adult to help them do something about it.

Why It's a Safety Issue, Not Just a Study Habit

What's at stake

Judgment. Critical thinking is what helps a child spot a scam, a deepfake or a chatbot that is flattering them.

Trust in the tool. A child who stops checking answers also stops questioning advice on friends, health and feelings.

The same app, many uses. The chatbot that does the maths homework is often the one they confide in later.

The skill that protects children from bad AI is the very skill that over-reliance on AI can wear down.

What to Do This Week

Ask them what they think Ask: "Does AI help you learn, or just finish?" Most kids already have an honest answer.
Try attempt first, AI second Agree that they try the problem alone first, then use AI to check or explain, never to start.
Make them the teacher Have your child explain an AI-assisted answer back to you. If they can't, they haven't learned it yet.
See the bigger picture Use an independent tool to see which AI apps your child uses and how often, beyond homework.

Your child already senses the risk. The best help you can give is a plan, not a ban.


Help your child stay the smartest one in the room.

Start a free 14-day Blaick trial — no credit card required.

Blaick gives you clear visibility into the AI tools your child uses, so you can guide them with facts, not guesses.

Start free trial →
Cancel anytime · Trusted by security-conscious families · Built on enterprise AI security