(Article too long? Scroll to the bottom for a summary)
Is teaching kids to code still valuable in a world where Artificial Intelligence is changing the landscape of software development? The short answer is: absolutely. I may be biased as the founder of theCoderSchool, but we absolutely believe it here. Here’s why teaching kids to code remains crucial, even as AI transforms this and every other industry.
In the future, the ability to use technology will be just as important as understanding how it works. Coding isn’t just about typing commands; it’s about understanding how computers think and function, so users can logically put together the necessary technical pieces to make something work.
For example, how does AI work? It uses data modeling along with machine learning to create an LLM which... is all very confusing, right? AI isn't magic, it's just (lots of) lines of code. While non-coders have a hard time understanding how data modeling might relate to machine learning (or what they even mean), advanced coders with the technical and logical experience are better equipped to understand how a new technology like AI works under the covers.
At theCoderSchool, we feel that learning to code is about the process, not the result. The process in this case means a fundamental understanding of how the code works. It's not just “coding a game” – it's “how does the code make the game work”? That's why we focus on the logic of the code itself, not just the end product – so our students have a technical foundation to understand and use whatever comes in the future, whether it's AI, quantum computing, or something else entirely.
I had a Chief Product Officer friend of mine recently comment that if he, someone not trained in coding, could vibe code something today, why would kids need to learn to code? Anyone can do it, he said, even without coding skills. I didn't think of my amazing answer until later (don't you hate when that happens?). But it occured to me that his very situation was the reason why more kids should learn to code.
That's because sure, a non-coder can code today with vibe-coding. But don't you think a trained coder could vibe-code something even BETTER? Of course they could. Because coders can better understand what a vibe-coding platform spits out (and why) and can steer the platform more effectively rather than blindly assuming the AI isn't hallucinating. If an AI platform tells you "ok, I'll add a heartbeat to your operating system to fire up an agent to send emails for you" (sound familiar, OpenClaw fans?), would a non-coder know how to interpret that? Is that even safe? Coders know the answer.
The future isn't going to be about "can you code" because anyone can use vibe-coding platforms. It's going to be about "can you code BETTER". And if you're trained as a coder, your answer will be a resounding YES.
Coding is, at its core, about solving problems. It's one of the only disclipines where learning it is literally solving logical problem after problem after problem until the student gets the result they want. How do you implement X? Think through the logic. How do you debug Y? Think through the logic.
It's also about constantly breaking a larger problem into a smaller problem. How many times have you been asked in an interview - how do you solve a complex problem? You break it into manageable pieces. Take a simple game like Flappy Bird. To code it, you break it down into pipes, birds, and scoring, working on each until you create the larger, complex product. This skill of simplifying a problem by dividing it is a crucial skill for any industry.
At theCoderSchool, our goal is to reiterate and practice problem-solving. We don't mind if code fails because that's when a student really needs to use their problem-solving skills to debug the issue. It's the process of building and debugging the code, and therefore understanding it, that's much more important. Anyone can ask chatGPT to build a Flappy Bird game – but only coders can go through the problem-solving steps to recreate it themselves.
AI makes coding more efficient and productive, and, undoubtedly, will do so even more in the future. Does that mean a company would need fewer coders on a particular team? Perhaps. But will there be less coding jobs in the market as a result? I say no.
Imagine you are the CEO of a company. You need to build 100 widgets, and before AI, that took 10 coders. Then, say with AI, each coder is twice as productive, meaning you only need 5 coders to build those same 100 widgets. As CEO, would you rather 1) fire 5 coders so you reduce costs and keep making 100 widgets? Or, 2) keep your costs at 10 coders, but build 200 widgets instead thus making twice the revenue?
I get it, that's a simplified example. But productivity means to produce more, not to use less resources to produce the same amount. Increased productivity isn't a reason to reduce the workforce. It's a reason to expand and be a more productive society overall. And it’s also a reason I believe there will be more tech jobs than ever in the future.
Remember the Disney movie Wall-E from the late 2000's? The movie imagines a future where AI has replaced humans for all tasks. No longer needing to work, people become fat slobs, sitting in chairs moved around by a computer and simply asking computers to do tasks for them. Without a need to develop skills, they all become lazy homogenous blobs staring at a screen all day without a care in the world (wait, my kids are doing that already...). It's an ideal utopia where humans don't need to lift a finger. But is it really?
There’s a great plot line from another Disney movie, The Incredibles, where the villain Syndrome tries to sell his super-weapons to the entire world so mere mortals can be powerful superheroes. Everyone wants to be powerful, but what happens when everyone is powerful? As the best line of the movie goes, “When everyone is super, no one will be.”
Ok, enough of the Disney references, but there’s an underlying lesson in both movies. If everyone simply relies on the power of AI, then does the human race become homogenous? Do we all become the same? Is no one super? I would wager that the human spirit won't allow it. Someone will always try to make better AI. Someone will use AI to invent something. Someone will build something that competes with AI itself. It's how humans are wired, and it's why AI can never completely take over human tasks. And it's why, IMHO, those with coding skills will be better positioned to differentiate themselves from those without.
When I founded this company over 10 years ago, I knew that technology would evolve, and coding would change with it. But I've always believed that teaching kids to think logically would never change because that's the basis of all technology. It's not about completing the project but rather thinking through it and understanding it. Understand it, and you can build it. As the legendary Steve Jobs once said, “Everyone should learn how to program a computer because it teaches you how to think.” Amen.
So carry on, I say. Keep on coding and keep working on those logical skills. It may well be what sets you apart from the rest, even in the future of AI. Like we've always said at theCoderSchool, Learn to Code. Change the World.®
