I Used AI to Learn AI — Is That Cheating?
I learned AI, finance, coding, and even debugging with the help of AI. But the interesting part is not what I learned; it is how I used AI to learn AI. I didn’t use AI just to ask for answers. Most of the time, I wanted to understand why something was happening in the first place. If my code gave me an error, I didn’t simply ask, “How do I fix this?” I asked, “Why did this error happen? What exactly is happening behind the code?” When I started learning finance, I didn’t stop at “What is inflation?” My next question was, “But why does inflation happen in the first place?” And when I started learning AI and used AI to learn AI, I slowly realised that getting an answer is easy. Understanding the reason behind that answer is where the actual learning happens.
That made me wonder: if I’m using AI to learn AI, am I actually learning, or am I just cheating? Because let’s be honest, students have always found creative ways to make studying easier. Pehle notes kisi topper se maangte the, phir YouTube pe “one shot in 30 minutes” dhoondhte the, aur ab ChatGPT ko bol dete hain, “Bhai please ye chapter samjha de, kal exam hai.” The technology changed, but the student behaviour is still very much alive. The difference is that AI can now sit with us at 2 AM, patiently explain the same thing five different ways, and somehow not get irritated when we ask, “But WHY?” for the seventh time. Honestly, if this were a real teacher, by the seventh “why” they might have started questioning their career choices.
The “Why” That Changed How I can use AI to learn AI
And that “why” is actually the reason I started seeing AI differently. I don’t really use AI as an answer machine. I use it more like that one senior didi who somehow knows everything and is willing to explain it to you without making you feel stupid for not knowing it. If I don’t understand something from my textbook, I can ask AI to explain it in simple language. If I still don’t understand it, I can ask for an example. If the example makes things even more confusing, I can say, “Okay, explain it like I’m five.” And if that still doesn’t work, well, apparently I am now three years old and we start again from the beginning. At this point, AI isn’t teaching me a university subject anymore; it is teaching me ABCD with a degree attached.
But this is exactly what I like about learning with AI. I don’t have to pretend that I understood something just because the textbook used complicated words and I nodded at the page for five minutes. I can stop and ask the most basic question possible. I can ask the same thing again. I can ask for another example. I can even say, “No, this explanation made it worse.” And AI will simply try again. That freedom to ask without feeling embarrassed is something I genuinely find useful as a student.
How AI Became My Coding bro
This became especially useful when I started coding. Coding has a very special way of making you feel like a genius for five minutes and then making you question every life decision you have ever made because of one tiny red error message. You write 50 lines of code, everything looks beautiful, you run it, and Python basically looks at you and says, “No.” No explanation. Just emotional damage.
When my Python code doesn’t work, I could simply paste it into AI and say, “Fix this.” And yes, it will probably fix it. But then what? Tomorrow I’ll make the same mistake again, and we’ll be back here like nothing happened. So instead, I started asking, “Why am I getting this error? Which part of my logic is wrong? What is Python actually doing here?” Sometimes I even tell AI not to give me the answer immediately and just give me a hint. Basically, I make AI do the explaining while I do the actual suffering required to become a programmer.
And honestly, debugging became much more interesting when I started doing this. Instead of seeing an error message as something that had personally attacked me, I started seeing it as a clue. AI could help me understand what the error meant, but I still had to connect that explanation with my own code. That difference matters because the next time a similar error appears, I don’t have to run to AI like, “bro, there is an emergency.” I have at least some idea of what went wrong.
Then Finance Entered the Chat(this is the main part where i learn how to i can use AI to learn Ai)
The same thing happened when I started learning finance. I could have memorised a definition of inflation and moved on, but that has never really worked for me. If someone tells me, “Inflation means the general increase in prices,” my brain immediately goes, “Okay, but WHY are prices increasing?” Then I ask why demand increases, why supply doesn’t always keep up, why petrol prices affect other things, why inflation happens in different countries, and suddenly one simple definition has turned into a full investigation.
At some point, a perfectly normal question like “What is inflation?” becomes a family tree of questions. “Why did this happen?” leads to “Okay, but why did THAT happen?” and then suddenly you’re three Wikipedia tabs deep wondering how the price of potatoes is connected to the economy of an entire country. But that is exactly when I feel like I’m actually learning. I’m not memorising a sentence just because someone wrote it in a textbook. I’m trying to understand the chain of events behind it.
And this is one of the biggest reasons I like using AI for learning. I can follow my curiosity instead of stopping at the first definition. If something doesn’t make sense, I don’t have to wait for the next class or search through ten different websites hoping someone explained it in a way my brain understands. I can simply ask again.
Is AI Making Students Lazy?
And I think this is where the whole “AI is making students lazy” argument gets a little complicated. Yes, AI can absolutely make students lazy. If I ask AI to write my entire assignment, copy it, submit it, and then forget everything about it, obviously I haven’t learned much. I have basically hired a very intelligent intern and then taken credit for the work. The intern did the project, the intern wrote the report, the intern solved the problems, and I just showed up for attendance. Very entrepreneurial of me, but not exactly education.
But if I use AI to understand something I couldn’t understand on my own, ask questions, challenge the explanation, solve problems myself, and use AI to find where I went wrong, then I don’t think that’s laziness. I think that’s simply using a new tool to learn differently.
The tool itself isn’t automatically making us lazy. The way we use the tool decides whether it helps us learn or helps us avoid learning. A calculator can help you calculate faster, but if you don’t know what calculation you need to perform, the calculator isn’t going to save you. AI is similar. It can make the process faster, but you still need to understand what you’re trying to do.
“Give Me the Answer” vs “Help Me Understand”
There is a huge difference between saying, “Give me the answer” and saying, “Help me understand how to find the answer.” The first one can make AI your replacement. The second one can make AI your learning partner. And honestly, I prefer the second one because there is something very satisfying about finally understanding a concept that looked completely horrible ten minutes ago.
You know that moment when you’re staring at a topic thinking, “What even is this?” and then suddenly it clicks? That moment is the actual win. Not the answer. The understanding. The answer is just the receipt. The actual purchase is knowing what you are doing.
Sometimes I even intentionally ask AI to not give me the answer. I’ll ask it to give me a hint, ask me questions, or tell me where my thinking went wrong. It feels slightly ridiculous because I’m literally sitting in front of a machine that can probably solve the problem in two seconds and I’m telling it, “No no, don’t tell me. Let me struggle.” But that struggle is useful. Otherwise, I’m just watching AI exercise while my brain sits on the sofa eating chips.
The “What If ChatGPT Disappeared Tomorrow?” Test
But there is one test I like to use whenever I wonder whether I’m actually learning from AI. I call it the “What if ChatGPT disappeared tomorrow?” test. Imagine you wake up tomorrow and AI is gone. No ChatGPT, no AI assistant, nothing. Could you still explain what you learned yesterday? Could you write a basic version of the code? Could you solve a similar problem? Could you explain why your solution works?
If yes, then congratulations, you probably learned something. If your first thought is, “Wait, but I need ChatGPT to run this,” then maybe we have a small problem. And if your second thought is, “Can I somehow download ChatGPT into my brain?” then congratulations again, we have reached the dependency stage.
This test is simple, but I think it tells us a lot. AI should ideally make you more capable over time. If every new problem makes you more dependent on AI, something is wrong. But if every explanation helps you solve the next problem a little more independently, then AI is actually doing what a good learning tool should do.
The Copy-Paste Trap
Because having working code doesn’t automatically mean you know how to code. AI can give you a beautiful solution with perfect formatting, impressive variable names, and comments that make you feel like you just became a senior developer. Then your professor asks, “Why did you use this function?” and suddenly you are looking at your own code like it was written by a mysterious stranger.
That’s the copy-paste trap. The code works, the assignment is submitted, but your brain is still buffering.
And this isn’t only about coding. The same thing can happen with essays, presentations, research, mathematics, and almost anything else. You can have a perfectly polished final answer without actually understanding the topic. The danger is that the result can look much smarter than the person who submitted it. And eventually, reality catches up. Exams, interviews, projects, or one very curious professor can expose the difference between having an answer and having knowledge.
So, How Should Students Actually Use AI?
This is why I don’t think the question should simply be “Should students use AI?” That question is already outdated. Students are going to use AI. The more useful question is “How should students use AI without replacing their own thinking?”
AI can explain a difficult topic, give examples, help debug code, create practice questions, challenge an idea, and even show you a completely different way of looking at a problem. But you still need to understand what it is telling you. AI should make the learning process easier, not make learning itself unnecessary.
For me, the best way to use AI is to make it do the things that support my thinking, not the things that completely replace it. I can ask for an explanation, but then I try to explain the concept back in my own words. I can ask for help with code, but then I read and understand every important part of the solution. I can ask for practice questions, but I still solve them myself. Basically, AI can sit beside me with the notes, but unfortunately, my brain still has to attend the class.
But Wait… AI Can Also Be Wrong
There is another important thing to remember: AI can be wrong. Very confidently wrong. And somehow it can make a mistake while sounding like it has just returned from giving a TED Talk.
That’s why blindly trusting AI is not AI literacy. If I’m learning with AI, I also need to question it, verify important information, test the code, compare explanations, and use my own judgement. AI can be incredibly useful, but it isn’t a magical machine that turns every sentence it generates into a fact. UNESCO has also highlighted the opportunities and risks of generative AI in education, including concerns around reliability, ethics, privacy, and academic integrity.
So if AI tells me something important, especially in areas like finance, technology, research, or academics, I don’t want to just say, “Okay, AI said it, so it must be true.” That’s basically replacing one textbook with another textbook that occasionally makes things up with confidence.
So, Is Using AI to Learn AI Cheating?
So, is using AI to learn AI cheating? I don’t think so. If I’m asking AI to explain a concept, helping me understand my mistake, giving me examples, testing my knowledge, or guiding me when I’m stuck, I am still doing the learning. But if I ask AI to do everything, copy the result, submit it, and then can’t explain what I submitted, then yes, something has gone wrong.
The problem isn’t that AI helped me. The problem is that I stopped participating in my own learning.
Academic rules can also differ between universities, professors, and assignments, so students should always follow the specific rules they’re given. But outside of that question of academic policy, I think the learning distinction is pretty simple: Did AI help you understand something, or did AI understand it for you?
My AI Is Basically That One Smart Senior bro
For me, AI works best when I treat it like a smart senior bro. I can go to him and say, “bro, ye samajh nahi aa raha.” he explains it. I ask another stupid question. he explains again. I ask, “But why?” he explains that too. Then I try something myself, make a mistake, come back, and ask what went wrong.
The only problem is that this bro has infinite patience, knows Python, knows finance, knows machine learning, can explain recursion at 2 AM, and never says, “Beta kal padhna, mujhe bhi sona hai.” Honestly, slightly suspicious.
But at the end of the day, I still have to sit for the exam. bro isn’t going to magically appear in the examination hall and whisper the answer into my ear. Sadly, technology has not reached that level yet. Although, considering how fast AI is developing, I’m not making any promises about the future.
I Don’t Want AI to Think For Me i want to use ai to learn ai
So yes, I used AI to learn AI. I used it to learn coding, debugging, finance, and a lot of things that once felt unnecessarily complicated. But I didn’t want AI to simply hand me answers. I wanted to know why those answers existed, why the problem existed, and how everything connected.
And I think that’s the difference between using AI as a shortcut and using AI as a learning tool. A shortcut gets you to the destination faster. A learning tool helps you understand the road you’re travelling on. If I reach the destination but have absolutely no idea how I got there, that’s not exactly a successful learning experience.
Maybe the goal isn’t to learn without AI anymore. Maybe the goal is to learn with AI without forgetting how to think without it.
Because AI might give you the answer in two seconds.
But knowing which question to ask in the first place?
That’s still your job.
Before I end this article, you might be wondering why I kept writing AI to learn AI again and again. Well, welcome to the glamorous world of blogging, where sometimes you write AI to learn AI because you genuinely mean it, and sometimes you write AI to learn AI because Google apparently likes knowing what your article is about. The funny thing is that while editing this article, I realised my main keyword, AI to learn AI, had only appeared four times. Four. After writing an entire article about AI to learn AI. Clearly, my brilliant blogger brain was not functioning at full capacity. In my defense, I’m still new to blogging and SEO, so some lessons arrive a little later than others. Right now, I’m also extremely sleepy, but I had to finish this because I made a promise to myself that I would publish something every Sunday. Unfortunately, by the time this article goes live, it will probably be Sunday night—scientifically close enough to Monday for some people. But for the sake of my consistency streak, my heart, and my sleep-deprived brain, I am officially counting it as Sunday. 😴✍️
Now I think we have covered the part on how you can use AI to learn AI
if you want to know more about AI, visit this page https://btawrites.com/what-is-ai/
UNESCO — Guidance for Generative AI in Education and Research
UNESCO ka official guidance hai on responsible, ethical and human-centred use of generative AI in education. Academic integrity aur AI-generated work ke issues ke liye particularly useful.
UNESCO — Guidance for Generative AI in Education and ResearchUNESCO — AI and Education: Protecting the Rights of Learners
AI ke educational benefits ke saath privacy, ethics, safety, equity aur human-centred learning ke risks discuss karta hai.
UNESCO — AI and Education: Protecting the Rights of LearnersEDUCAUSE — 2025 Students and Technology Survey
Ye directly students ke generative AI use ko cover karta hai, including explaining problems without getting answers, creating study materials, learning workforce skills, developing AI literacy, aur AI ke risks like over-reliance and shallow learning. Tumhare article ke liye VERY relevant source hai.
EDUCAUSE — 2025 Students and Technology SurveyEDUCAUSE — 2025 Students and Technology Report
Higher education mein students, technology aur generative AI ke changing relationship par research/report.
EDUCAUSE — 2025 Students and Technology Report
