By simplifying the task of coding, artificial intelligence has increased the importance of learning computer science.

By Hadi Partovi
Code.org cofounder and CEO

Another year, another viral tweet argues against “learning to code.” The latest one comes with cherry-picked data implying that art history majors outperform computer scientists in labor markets. Yet the underlying data from the Federal Reserve Bank of New York prove the opposite. Art history grads have low unemployment—just 3%—because they are actually underemployed: 47% work in low-wage jobs that require no college degree and earn almost half the median wage of computer scientists. Computer science grads, meanwhile, have the highest median wage and the fifth-lowest underemployment across all majors.

Skeptics are right to point out that AI will automate the busywork of coding. Software jobs will change just as they have every decade, as programming has become easier and easier since the days of punch cards and assembly language. But we shouldn’t be confused by the short-term shifts of some employers in some industries. Simplified coding hasn’t decreased the value of computer science. Rather, it has increased, and the demand for computer science will grow over time.

But here’s the thing: Debating the employability of computer scientists ignores the most important reason to teach computer science.

The value of computer science has never been solely about tech careers. It’s about equipping students with the foundation to navigate an increasingly digital world, regardless of their passion. Computer science is a liberal art.

Critics ask, “Why learn computer science when AI can code?” But nobody questions learning reading, writing, math, science and history, although AI does all these things as well. AI will certainly change the nature of coding, but by lowering the bar for software engineering, it makes foundational computer science more important than ever.

Andrew Ng, one of the world’s foremost AI leaders, puts it bluntly: “Some people today are discouraging others from learning programming on the grounds AI will automate it. This advice will be seen as some of the worst career advice ever given.” As coding becomes easier, he argues, more people should learn it, not fewer. Understanding the “language of software,” as Ng calls it, is how people in every profession will get better results from AI. It’s how you learn to tell a computer exactly what you want it to do.

This sentiment is echoed across the tech industry: by OpenAI’s Sam Altman, Google’s Demis Hassabis and hundreds of CEOs including Microsoft’s Satya Nadella who have called for computer science and AI education to be a required part of every K-12 student’s learning. The goal isn’t to turn every kid into a coder but to give every student agency in a world where AI is no longer optional. 

In the words of Steve Jobs, “Everybody should learn how to program a computer…because it teaches you how to think.” That’s even more true today, as AI reshapes every industry—from finance and healthcare to journalism, manufacturing and the arts.

Computer science teaches problem-solving, data literacy, ethical decision-making and how to design complex systems. It empowers students not just to use technology but to understand and shape it. That kind of adaptability is invaluable in an economy where the required skills for most jobs are projected to shift 65% by 2030. 

This is especially critical in K-12. Early exposure to computer science builds problem-solving, critical reasoning and technological fluency—skills that matter even for students who never pursue a career in tech. One high school computer science course alone increases earnings 8% and employment likelihood 3% for all students, regardless of what field they enter.

So yes, let’s correct the record on job stats. But more importantly, let’s stop defining the value of computer science by employment charts alone. In the age of AI, computer science isn’t just a career path. It’s a new kind of literacy—one every student deserves.