Should You Learn Coding in 2026, or Will AI Handle Everything?
Jensen Huang recently stated that AI will do most of the work in the future, and people won’t need to learn coding; coding will become obsolete, meaning there will be no need for coding. Even Elon Musk said in 2022 that AGI (Artificial General Intelligence) would arrive by the end of that year, and in 2025 and 2026, AGI still hasn’t arrived.
Now, if anyone asks me today whether they should learn to code or not, my counter-question to them is: “If you don’t learn coding let’s say you were in the coding field, but now you think AI has arrived and will do all the work what will you do then? Tell me, if not coding, then what?”
Why Coding Can Never Truly Be Obsolete
Many people’s answer is, “I’ll go into data analytics,” or “I’ll go into data collection,” or they talk about a completely different profession, like cooking.
Look, you need to understand one thing here: Almost all professions are being automated with the help of AI.
When big personalities make such statements, like Jensen Huang’s claim that coding will become obsolete, there’s a hidden motive. It’s their way of saying that they are very confident in their company. Let’s take Jensen Huang’s example. He is the CEO of Nvidia. He has a great love for Nvidia and what they do at Nvidia. He is super-optimistic about AI because they are creating GPUs and supporting AI companies through investments.
So, he is overly optimistic. And sometimes, being overly optimistic isn’t necessarily wrong for someone, because if I don’t believe I can do something, how will I do it? Similarly, Jensen Huang says, “How can it not happen? We will bring AGI, come what may.” Now, if someone doesn’t believe that AGI will ever come, how will they bring it?
So, all these people motivate themselves and tell the world that all these things are possible. They remain overly optimistic, but you shouldn’t get demotivated by their statements. At least, this is what I feel.
The Power of Knowing How to Code in the AI Era
It is easy for Jensen Huang to say that coding will become obsolete and AI will do everything. But this statement, in itself, is a very flawed statement. When will it be obsolete, and what does obsolete mean? Does it mean coding will end, and AI will do it? AI will also be doing coding, right? Will it do 100% of the coding? How? Humans must always remain in the loop. You need to increase your productivity and incorporate AI into your workflow as much as possible.
I personally use many tools to increase my productivity. If you don’t learn basic programming, how will you be able to write good prompts?
Let’s say you are creating a website using AI. I am not portraying myself as a great coder here; I assume you are a beginner, and I am an intermediate developer. Let’s say I compete with you—you are sitting on one side, and I am on the other. We both have to create an e-commerce website, and whoever finishes first wins, and whoever makes the better one wins. We both have all the AI tools available in the world. Tell me, can you win against me?
Now, pay attention. I’m not claiming to be the best coder in the world, but if you’re a beginner, can you win against me? Or if you’ve never coded in your life, tell me, can you win against me?
The answer is no. You can’t win against me, at least not when you are a beginner. Why? Because my prompts will look something like:
- “Create a middleware that modifies the request object in this way.”
- “Create an endpoint.”
- “Create a document in MongoDB.”
If you don’t even know what a document is, what fields a document has, what a schema is, or what an ID is, how will your prompts be? Your prompts will be: “Create an e-commerce website for me.” “Make the design of this website look good.”
My prompts will look like: “Reduce the opacity.” “Make the object-cover property uniform across all sections.”
So, when I write such sophisticated prompts, my website will obviously be better.
AI’s Current Capabilities and Human Advantage
If you are starting from zero today and you think that AI can do a lot, AI will definitely be able to do a lot, because AI will stay at the level it is at. But if you are at zero today, you will go from zero to, say, 10, then to 15, then to 20.
Let’s say today AI scores 40 out of 100 (if we talk about marks, including everything: AI’s incapability, turning on the computer, or getting into a server, you have to prompt it for everything, and sometimes you see hallucinations).
You will come to 15, then 20, then 30, and you have the capability to go up to 70, but AI is sitting at 40 and will stay at 40, max to max, it might reach 45.
Today, most AI startups are overvalued. People are blindly investing money in AI. I will share a report with you that you will enjoy, and you will agree that a lot of hype has been created around AI. Yes, there is a lot of truth in AI. We have seen ChatGPT; it is useful, it works. We have been using it for so long now, and somewhere we feel that we are using ChatGPT as much as we can in our daily lives.
The AI Bubble: Lessons from the Dot-Com Era
Derek Thompson has written a report that is going very viral these days, and I want to share this report with you. He discussed whether the AI bubble is going to pop or not, just like the dot-com bubble popped.
If you don’t know about the dot-com bubble, in 2000, when the internet came, people were madly investing money in the internet. People thought the internet was the “next big thing.” There was a website called Pets.com in 2000, in which people were blindly investing money because it supplied pet supplies. People thought, “Now, everything will be done by this, all the shops will close, everything will go online.” And that startup shut down.
Why You Should Still Learn Coding in 2026
The investors who put money into that company lost everything because they did not have a realistic way to make a profit. They were spending excessively on various things based on investor money, and their road to profitability was not clear. That’s why that company shut down. But does that mean the internet is useless? No.
Similarly, the AI scenario is that AI is not useless. AI is, in fact, very useful, and you and I will use it in the coming time. But the adoption will be steady, slowly, and gradually, and we will adopt it and increase our productivity. It will not happen that one day someone creates an AI that means nobody needs to do anything; the AI will do everything.
So, is this an AI bubble? Only time will tell; I am nobody to predict. But look at the patterns. Derek also says the same thing here. I will link the report in the description so you can read it in detail, but I’m telling you the key takeaways from it.
One thing he said here is that today, investors are pouring money into AI startups:
What Happens If You Don’t Learn Coding
If you look here, the capital expenditure of these big companies Amazon, Meta, Google, Microsoft, Oracle in 2018 was below $100 billion, around $80-90 billion. Now, in 2026, the same capital expenditure has reached $500 billion in 2026-27. Do you understand what this means? It means these people are spending $500 billion, while in 2018, they were spending $70-80 billion. Now they are spending $500 billion, and how much output is coming from it? That is, how much are they selling in goods and services? It’s only $12 billion! And they are spending $500 billion! They will make this money in the future.
Now, you are wondering what the problem is here. This is exactly what happened during the dot-com bubble.
Now, whether AI is a bubble or not, time will tell. But look at the patterns.
Here is a startup named Thinking Machines, created by OpenAI’s Mira Murati. What does this startup do? Nobody knows! It’s just a company. Because Mira Murati worked at OpenAI, people started investing money with the hope that she will do something big. And its valuation has reached $10 billion! Do you understand? The company has done nothing, but it has a valuation of $10 billion. Why? “Oh, Mira Murati worked at OpenAI, so Mira Murati will create something good.” Now, this doesn’t make any sense to me, but it makes sense to investors, due to which investors have given it a valuation of $10 billion, and they have also raised a funding of $2 billion.
So, people are blindly investing money in AI, and in the coming time, if they realize that its output is not going to come in the near term (if it comes in the next 20 years, then we’ll see; if the money spent today comes in the next 10 years, then we’ll see), suddenly this AI bubble will burst.
Investors will quickly pull out their money at low valuations and cut their losses, due to which this funding they are receiving will stop, and they won’t be able to make this massive capital expenditure. There are many other details in this article. You must read it; you will enjoy it. If you have even a little time to read, then definitely read it. I will link it in the description. It has been curated very well. Full credits to Derek Thompson. I enjoyed reading the article.
So, my point here is simply that AI is a magical autocomplete. Many people will say, “Why are you defending the fact that learning to code is still worth it?” I am not defending; I am actively doubling down on my learnings. I am learning more because today I feel that where I used to learn one technology and switch between different languages and write code, for example, today I am doing PHP, tomorrow C, the day after C++.
I am talking only about myself. I am not advising you to learn three languages. Today, it has become so easy for me to switch from C to C++, and from C++ to PHP (I am talking about real-world, big projects). This has become very easy for me. So, I say, “I will learn more technologies.” It has become very easy for me all thanks to AI, both in terms of learning and implementation.
So, learning the basics of anything is worth it, and I would say it’s more worth it than before. Because today, if you learn even a little Python, you will become a coder.
When ChatGPT was launched, and up until today, ChatGPT has made many improvements. Back then, ChatGPT used to make big mistakes. I still remember it used to give wrong code with great confidence and wrong information related to coding with great confidence. Today, this has reduced significantly. This has reduced significantly even with AI for audio, AI for images, and AI for videos. They have also advanced significantly.
Personally, I used Sora 2. It was good, but Sora 2 is not good for my use case at all. What was my use case? I wanted to create some explainer animations, and Sora 2 could not create those explainer animations for me. And even today, I do these things the traditional way. We are editing videos the traditional way today. Those things are not at that level when it comes to explainer videos. But yes, somewhere, AI for video is giving exceptionally well outputs.
So, things are changing slowly. Adoption is happening. AI is doing very well. But that doesn’t mean you shouldn’t code.
Now tell me, if you don’t code, what will you do? You are telling me, “Since AI will code in the future, I won’t code.” So, what will you do? “I’ll do something else.” But that “something else” will also be done by AI, right? So, what will you do? Sit idle? This is my genuine question, and if you have the answer, tell me in the comments section. What will you do? Sit idle?
You can name any field. Many people say farming. I will tell you how much AI research is going on related to farming. People say, “Do farming. There is no AI in farming.” But once you land, even farming can be automated. Look, everything is possible. Even hair can be cut by AI. Everything can be done by AI. If we are only talking about possibility, if you say that AGI will come and AI will code better than humans, then I can also say that AI will do farming. Then AI will do everything.
This is a very negative mindset. We should adopt a positive mindset and move forward using AI tools.
I tell you that even if you start today and work hard for a year in coding, in the next one or two years, you will make a lot of money in coding. A lot of money will be made, all thanks to AI. And if you learn AI-related things like data science and machine learning, your learning will go to the next level, and you can definitely convert that learning into money.
So, the conclusion of this is what I want to give you: Just start coding. Start any kind of coding. If you want to start with Python, do Python. If you are in your first year and want to start with C language, start with C language. You can definitely start with C++ as well. If you are a web developer, you can also start JavaScript as your first programming language after HTML and CSS.
But one thing I am against is sitting idle and crying that “AI will eat jobs, and there is nothing left in the future.”
The Smart Way Forward
In my opinion, and in the opinion of many, AI will progress a lot in the coming time. But who will drive the progress? Coders. What do people at OpenAI do? They code. How are OpenAI’s models built? They are built by coding.
In my opinion, people should not run away from the basics. If you are a good developer, and especially a developer who knows how to code from the AI side, and you say, “I will do that work from the AI side that a human might take 10 hours to do,” then your identity as an AI-assistant web developer or a web developer who uses AI will be greatly established, and you will be able to flaunt this on your resume.
If you have built 10 apps, if you say, “I have built a clone of Facebook, I have built a clone of Instagram, I have built a clone of all the big services, and it runs very well, and I also have a good user base, and you also know the basics that this works well because we are doing this and this,” and where you would have taken time as a senior developer, you saved that time with the help of AI.
Do such work, instead of thinking, “I won’t code at all.” Look, if you become foolish and get convinced that you shouldn’t code, the good news for the one who codes is that people are stepping back from coding, and a place is being made for them. Already, AI has come for them, which is helping them in learning and implementations. Along with this, if the competition also reduces, it’s the icing on the cake.
Conclusion
So, in conclusion, I will say: Work hard. You will move forward by working hard. Start today and waste as little time as possible. Time is the most valuable asset you have. AI things will come and go. If a new product comes, definitely try it. If it fits into your work, fit it in. Start the adoption of AI too, and don’t demotivate yourself by saying that “AI will end this thing, AI will end that thing.”
