For years, young people were told some version of the same thing: go to college, get a degree, and that is how you build a successful career. I heard it. My friends heard it. A lot of us followed that path.
I studied computer programming in college, and I had friends who took the same courses I did and earned the same degree. Some of them never worked a single day in the software industry.
That always stayed with me, not because there was anything wrong with the degree, but because for some people, getting the degree had become the goal. It was something you were supposed to do. Get the qualification, check the box, and somehow the career would follow.
Lately, I have been wondering if we are starting to do the same thing with AI.
Learn AI
You hear the advice everywhere now: learn AI, learn how to use ChatGPT, learn prompting, learn automation, learn agents, learn how to code with AI.
To be clear, I think people absolutely should understand AI. I use it every day. It has changed the way I build software, research problems, think through ideas, and get work done.
What concerns me is that we may be turning “learn AI” into a destination.
Learn AI to do what?
That is the question I think gets skipped. If someone learns every major AI tool but has no real understanding of a business problem, a profession, a craft, or an industry, what exactly have they prepared themselves to do?
Knowing the tool is not the same thing as knowing what to do with it.
I See This in Software
AI can write code, and sometimes it can write very good code. It can help me troubleshoot problems, explain unfamiliar code, suggest different approaches, generate documentation, and move much faster than I could before.
But that does not mean the person using it no longer needs to understand software. Someone still has to know what should be built, how the pieces fit together, and whether the solution actually makes sense. Someone has to notice when the AI gives you something that looks perfectly reasonable but is completely wrong for the application.
Someone also has to talk to the customer and understand what problem they are actually trying to solve. And when the software does not work, someone has to take responsibility for fixing it.
AI can help you do the work. It does not automatically teach you what good work looks like.
Maybe Expertise Matters More, Not Less
A lot of people are worried that AI will make their knowledge less valuable. In some cases, that may happen. But I also think the opposite can be true.
The better AI gets at producing things, the more important it becomes to recognize whether what it produced is any good.
A graphic designer can generate twenty concepts in minutes, but someone still has to know which one actually communicates the message. A programmer can generate hundreds of lines of code, but someone still has to know whether the architecture makes sense. A business owner can ask AI for a strategy, but someone still has to understand the business well enough to know whether that strategy is realistic.
That kind of judgment comes from somewhere. It comes from experience, practice, mistakes, and actually doing the work. AI can speed up a lot of that work. I do not think it removes the need to understand it.
What About AI Taking Our Jobs?
This is also why I think we may be approaching the jobs conversation the wrong way. People understandably ask whether AI will replace programmers, designers, accountants, marketers, and many other professions.
I am not going to pretend that AI will not eliminate some jobs or reduce the amount of human work needed in certain areas. It will.
But I think there is another question that may be more useful.
How is AI changing the work I do, and how do I become better because of it?
That creates a very different mindset. If you are a programmer, learn what AI changes about programming. If you are a designer, learn what it changes about design. If you are an accountant, learn what it changes about accounting.
And if you are still figuring out what you want to do, build real skills first, then understand how AI is changing that field. That seems much more useful to me than simply trying to become “an AI person.”
Maybe We Are Giving People the Wrong Advice
I do not think the lesson is that people should stop going to college. I definitely do not think the lesson is that people should stop learning AI.
The lesson I take from my own experience is simpler: do not confuse the qualification with the career, and do not confuse the tool with the skill.
A degree can be valuable. AI can be incredibly valuable. But neither one automatically gives you something useful to do with it.
So maybe instead of telling young people simply to “learn AI,” we should tell them something a little harder. Learn how to solve problems. Become good at something. Understand people. Understand the industry you want to work in. Develop judgment. Learn what good work looks like.
Then learn how AI can make you better at it.
Become good at something that matters. Then learn how AI can help you do it better.
Years ago, we told people to get a degree. Today, we are telling people to learn AI. Maybe the better advice is to stop treating either one as the destination and focus instead on what we can actually do with what we learn.