بسم الله الرحمن الرحيم.
It is impossible to deny that Artificial Intelligence (AI) has become one of the most talked-about subjects in the world today; everyone uses it, and nobody stops talking about it and its endless updates. Yet amid all this excitement, we find two extremist camps: one believes AI will take over the world, solve every problem, and render us all useless; the other dismisses it entirely.
We do not belong to a camp that opposes everything for the sake of opposition, but we believe the balanced view is the right one. We are not here to tell you “what will happen” in the future — that is a question no one can answer — but what we know for certain is that the real problem today lies not in the tool itself, but in how we use it. In short: we simply do not handle AI well.
In this article, we will cover how to use AI correctly, whether as a student or at work — because the subject is, in fact, completely different between the two cases.
Before talking about how to use it, we must first talk about how not to use it.
1. Why You Should Not Ask AI to Solve Your School Project
The first thing we focus on in programming is problem solving — giving ourselves time to think through a solution, perhaps a full hour or two.
These training problems are “artificial” by nature, even though their answers are freely available online. So why do we spend an hour or two thinking about them? Why “waste” time on them instead of cloning someone else’s ready-made repo and declaring the project done?
You surely know the answer. If we did that, we would learn nothing. More precisely: we would never wrestle with the problem, and therefore we would lose the learning that comes from that struggle and search.
Merely asking AI to solve the problem adds nothing to your skills; it is simply asking someone else to do the hard work in your place.
The Gym Analogy
Think of it like going to the gym and asking a robot — or another person — to lift the dumbbells for you, then walking out announcing that the dumbbells were successfully lifted, while you gained nothing.
Can you name a single activity that fulfills the purpose of the gym other than lifting weights? Simply put: there is none. So why do we spend time lifting dumbbells if no other activity requires it? Because the dumbbells are merely tools we use to achieve the bigger goal: being stronger overall.
Studying works exactly the same way. The assignments you are asked to solve at school or university are the “dumbbells”: not real, but designed to give you exercise, so that when you reach real work you have the strength to get it done.
And because these problems are artificial, AI solves them easily. But do not be fooled: its ability to solve your schoolwork does not mean it can solve the real problems you will face on the job (not yet, anyway). If it were that easy, developer salaries would not have stayed high; if AI truly solved real-world problems, developers would be earning minimum wage.
In short: never ask AI to solve your programming project end to end. It may succeed — but you gain nothing. The point of the project was never the solution; it was developing the way you think.
2. How to Use AI as a Student
The first and most important rule: if your teacher or professor has banned AI, then it is banned — period. No debate.
If its use is allowed, however, treat it as a private tutor who knows many things reasonably well. But remember that this tutor makes mistakes and sometimes gives bad advice, so everything you learn from it must pass through critical thinking.
Here are the right ways to use it as a student:
- When stuck: when you are blocked on one small part of your project, ask about that specific part only.
- Recalling syntax: if you forget a particular piece of syntax, ask it directly.
- The missing piece: ask for the idiomatic way to write, say, a loop that removes elements from an array, or about a specific operator and how to use it.
- Speed: when you have small questions about a specific language or library, ask away; small, well-scoped questions are excellent — and far faster than traditional searching.
- Ask for hints, not solutions: this is where AI proves genuinely excellent. Give it code containing a function and say:
“There is a bug in this code [describe the problem]. But don’t tell me the answer and don’t give me the corrected code. Give me only hints about how to think about the problem so I can find the fix myself.”
- Compare and improve: once your project is finished and fully working, you can ask it for an alternative approach to compare against, or request improvement suggestions. But always keep in mind that such improvements sometimes add no real value — keep your critical eye on.
3. How Experts Use AI at Work
At work, the story is completely different.
Junior developers solve problems through logical reasoning alone, while experts — in addition to logical reasoning — rely on pattern recognition. As experts, they have a better grasp of the building blocks a problem is made of: they recognize the pieces they need and reason out how to assemble them together.
In other words, experts are distinctly better at two stages — understanding & planning — which are the essence of programming (though they are certainly better at every stage).
And here is where AI comes in: it helps experts write the code for those building blocks. They might ask it, for example:
“I need to filter these results for anything matching this regular expression.”
They then have it implement that unit, verify the correctness of the code with their own expertise, adjust it to fit their needs, and move on. Even with cases they do not know beforehand, this approach helps them get tasks done — but they still need their experience to know when they must go learn more.
AI is also very useful for proofs-of-concept and rapid prototyping, because the code in these cases is disposable and will be discarded later.
Summary
For these reasons, as a student, do not try to use AI like a seasoned developer; the code will come out full of bugs, you will not know how to fix them, and worst of all you will never build your skills.
As you gain more experience, you can lean on it to write large amounts of boilerplate code whose logic you already understand.
The discussion is open on this topic — if you hold a different view, do not hesitate to share it.
And do not forget us in your prayers, for me, my parents, and our struggling brothers across the world. Peace be upon you.