“No context, no truth”
Artificial Intelligence is no longer a futuristic concept; it’s a real-time job profile shredder. And in this new field of professional extermination, programmers are on the front line. The myth was that they were untouchable. The myth has just died.
A recent report published by ComputerHoy summarizes a thesis that is being repeated by increasingly authoritative voices: there are two types of programmers. Those who will lose their jobs because of AI, and those who will keep (or gain) one thanks to it. The line between the two is neither ethical nor technical: it’s strategic. In other words, it has more to do with forward-thinking than with lines of code.
“Copy-paste developers”: replacement fodder
On the left side of the ring are those who “know how to program,” but their greatest skill is retraining. They make a living by copying and pasting other people’s scripts, standard solutions, Stack Overflow, and templates. They don’t understand why they work, but they work. Or they did… until GPT, Copilot, or any halfway decent LLM came along.
These profiles are the first to fall. And with good reason: if your job can be done by a machine in less than 30 seconds with better documentation and fewer errors, it’s not a job, it’s an automatable routine. And like any automatable routine in the 21st century, it’s doomed to extinction.
The Architects of the Algorithm: Allies of the Beast
On the other side are the real engineers. Those who understand the principles behind the frameworks. Those who design solutions, not just execute them. And, above all, those who are learning to use AI as an assistant, not a replacement.
These programmers use GitHub Copilot as a copilot, not a driver. They understand that generative models can suggest, but not decide. And they know that an AI’s reasoning errors can be worse—and more dangerous—than a junior’s syntax errors. Their advantage lies not in speed, but in comprehension.
Code is language, not just instructions
A common misconception is that programming is simply translating ideas into machine languages. But programming is writing. And like all writing, it has context, intention, and style. AI can help you write, but if you don’t know what you’re writing, you’re still going to fall apart.
Therefore, the programmers who survive won’t be the fastest, but those who think best. Those who understand systems. Those who learn to formulate problems well. Because if AI does anything well, it’s solving… what it formulates well. Ergo: thinking well will be more important than writing well.
Programmers aren’t being replaced by GPT. They’re being replaced by programmers who use GPT better than they do. AI didn’t come to kill professions, it came to change them. It came to reward adaptability and punish complacency.
The most ironic thing is that many of those who mocked bank tellers, translators, or graphic designers now see the scythe touching their necks. There’s no karma. There’s logic. And the logic of the new world is clear: “Don’t compete against AI. Use it. Or you’ll be used by it.”
Chronicle of a Foretold Job Apocalypse
In the new digital order, code won’t save you. Only knowing why you’re writing it will.
While Silicon Valley gurus preach a utopian world where AI frees up time for “human creativity,” what’s actually happening is a systematic pruning of unnecessary job profiles. Artificial intelligence isn’t revolutionizing programming: it’s debugging it. And as in any debugging process, the first victims aren’t the useless, but the unnecessary.
The phrase circulating in digital corridors—“there are two types of programmers: those who will be replaced by AI and those who will have jobs thanks to it”—is not a metaphor. It’s a professional obituary.
AI doesn’t program for you. AI completes what you start. It suggests, predicts, recommends. But it doesn’t think. And that’s what many programmers forgot to do long ago: think. Instead of understanding structures, they apply recipes. Instead of designing, they assemble. They’re copywriters with mechanical keyboards.
And just as an AI doesn’t need to know what a metaphor is to imitate a poem, it doesn’t need to know what a pointer is to complete a function in C. But therein lies the dilemma: if you don’t know either, then what’s the difference between you and the model?
Right now, there are thousands—perhaps millions—of developers celebrating that GitHub Copilot “saves” them time. What they don’t know is that it’s also taking away their value. Every line of code an AI can write for you is one less line of code that justifies your salary.
It’s not about writing faster, but about knowing what’s worth writing. And that’s not dictated by Python or JavaScript. It’s dictated by the context, the architecture, the vision. In other words, it’s dictated by someone who thinks like an engineer, not a technician.
You don’t have to be a mystic to understand this: the only programmers who will have jobs in five years will be those who understand how AI works, not just how to use it. The difference between generating prompts and designing models is the same as between driving a car and building an engine. The former can be many. The latter, very few.
And if AI eventually replaces those who design it today—which, spoiler alert, is still a long way off—then the game won’t even be about programming. It will be about governance, algorithmic ethics, data interpretation, and automated decision architecture.
Many celebrated when AI started writing essays, painting pictures, or composing songs. “Art will finally be accessible,” they said. Today, when Copilot generates better code than the average computer science student, there’s not as much enthusiasm. Because the art belonged to others. But the code belonged to them. Or so they thought.
Code, dear friends, no longer makes you special. It’s no longer a passport to technological paradise. Today, writing code is as common as writing emails. The difference, again, is in the purpose. And AI still doesn’t understand that. But neither do you, if you continue to think that programming is an end and not a means.
We’re not facing the end of the programmer’s job. We’re facing the end of the programmer who doesn’t know why he works. And that species isn’t going to disappear because of AI. It’s going to disappear through inertia.
English 
Spanish 