← Back to the Research Program
AI & PeopleElementary researchExample brief
Do My Classmates Trust the AI's Answer?
The research question
When an AI gives an answer, do kids in my class believe it — even when it is wrong?
Abstract
I showed classmates AI answers, some right and some wrong, and asked if they believed each one. Many believed the wrong answers because the AI sounded sure.
Background
AI assistants always sound confident, even when they make mistakes. I wanted to know whether that confidence fools people.
What I did
I showed 12 classmates six AI answers — three correct and three wrong — and asked them to mark which ones they trusted.
What I found
Most classmates trusted the confident wrong answers. Only a few said they would check first.
What's next
I want to teach a short lesson on checking AI answers and then test whether trust changes.
Takeaway
An AI sounding sure does not mean it is right — checking is a habit worth teaching.