We use cookies. To find out more, read our Privacy and cookie policy.
OK, don't show again
We use cookies. To find out more, read our Privacy and cookie policy
Ok, don't show again
Leave your phone number and we will contact you!
Or you can call us:
+1 (516) 942-2021
By clicking the button you agree to our Privacy Policy
AI & Education
AI in education. Speed ≠ learning: why "less effort" often means "less retained"
AI & Education
AI in education. Speed ≠ learning: why "less effort" often means "less retained"
3 min read
3 min read
AI can help students finish faster — and remember less.

That is the uncomfortable part of the education debate that I think many people still underestimate.

In a study by AI researcher Andrew Barkley, 120 students were split into two groups. One group was allowed to use ChatGPT while preparing an assignment on artificial intelligence. The other studied the traditional way, without AI support.
Forty-five days later, they were unexpectedly tested.

The results were telling:

The ChatGPT group scored an average of 5.75 out of 10.
The non-AI group scored 6.85 out of 10.

At the same time, the AI-supported group finished much faster: 3.2 hours on average, versus 5.8 hours without AI.

That is what makes this interesting.
AI clearly helped students complete the task faster.
But it did not help them retain the material better.

And that forces a more serious question: is AI improving learning, or just reducing the friction of getting through it?

The explanation behind this is also familiar. If part of the mental work is outsourced to a tool, the depth of processing can fall with it. Less effort often feels better in the moment. But effort is also what helps knowledge stay in memory.

That is why I do not see this as an argument against AI in education.
I see it as an argument against lazy instructional design.

If AI is introduced without clear boundaries, students may get better at finishing assignments without getting better at understanding the material.

My view is simple: in education, speed is not the same thing as learning.
How are people handling this in practice now: AI for explanation and feedback, but not for first-pass study? Or is that boundary still blurry?
Explore More Insights from ITUniversum: