AI and Learning: 3 Limitations You Need to Know About
What science tells us about being careful when using AI as a learning tool
Lately, have you caught yourself turning to ChatGPT or Claude to learn something new? Maybe you're trying to grasp a complex concept, pick up a new programming language, or just understand something about Topic X. It seems AI has become our always-available teacher, ready to help 24/7.
I was thinking about this recently while reading a fascinating article in Harvard Business Publishing Education called "3 Critical Problems Gen AI Poses for Learning." Written by Jared Cooney Horvath – a neuroscientist and educator – it really got me thinking about how we use AI in our learning process.
Before we go any further, I want to clarify something important: there's a huge difference between studying and learning. Studying is what you do when you read, take notes, or complete exercises. Learning happens when that knowledge actually sticks in your brain and you can use it. AI is fantastic for helping you study (with summaries, Q&As, practice exercises), but how effective is it at helping you actually learn?
According to science, there are three important limitations we should keep in mind:
1. Empathy matters (a lot)
Imagine you're learning to play guitar. Which approach do you think would work better?
Option A: Super well-explained video tutorials
Option B: A teacher who notices when you're frustrated and knows exactly how to motivate you
Research shows that Option B wins by a landslide. Why? Because when you learn with someone who can truly empathize with you, your brain releases a hormone called oxytocin. It's like your brain and your teacher's brain sync up, making the learning process dramatically more effective.
AI, no matter how sophisticated it gets, has no oxytocin and no brain to synchronize with yours. It might simulate empathy convincingly, but it simply can't create that real chemical connection that makes learning stick at a deeper level.
2. Knowledge is not just access to information
"Why memorize anything when Google has everything?" You've probably heard this before. Now with AI, it's tempting to think: "Why bother learning when I can just ask ChatGPT?" (or Claude 😉)
But here's the problem: your brain isn't a computer that simply retrieves information. It's more like a chef who needs ingredients (knowledge) stored in their kitchen (memory) to create new dishes (ideas).
Even when you're not consciously focusing on a problem, your brain keeps working on it, connecting information you've already stored. That's why you suddenly get brilliant ideas in the shower – your brain has been quietly cooking with the ingredients it has on hand.
Here's the catch: this process only works with information that's already in your memory. It can't use information that's "stored in the cloud" or in ChatGPT. It's like trying to cook dinner with ingredients sitting on supermarket shelves instead of in your own kitchen – it simply doesn't work.
3. Multitasking is the secret enemy
"I'm learning SQL with YouTube tutorials while checking my email and chatting with ChatGPT about questions I have."
If this sounds familiar (I do it too 😉), I've got some bad news. Studies show that people typically manage only about 6 minutes of focused learning before getting distracted by something else on their computer. And when we bring laptops to class? We spend a whopping 38 minutes of every hour doing things completely unrelated to the lesson.
The problem isn't AI itself—it's that we access AI through devices that are essentially sophisticated distraction machines.
So what can we do?
I'm not suggesting you abandon AI for learning. On the contrary, it can be an incredibly helpful tool, especially when you already have a solid foundation in the subject. Think of it like using a calculator—incredibly useful if you already understand math, but not such a great idea if you're just learning how to add.
Here are some practical recommendations:
Use AI as a complement, not as a replacement for human teachers or mentors.
Make sure you truly understand and memorize the fundamental concepts before relying on AI.
Create a dedicated learning space where you can minimize digital distractions.
Next time you sit down to learn something new with AI's help, keep these limitations in mind. Not to discourage you, but to help you use these powerful tools in the most effective way possible.
See you soon,
G
Hey! I'm Germán, and I write about AI in both English and Spanish. This article was first published in Spanish in my newsletter AprendiendoIA, and I've adapted it for my English-speaking friends at My AI Journey. My mission is simple: helping you understand and leverage AI, regardless of your technical background or preferred language. See you in the next one!