Be nice to your parents. There's a decent chance you'll be living with them longer than either of you expected.
For twenty years people have predicted the death of college. The argument was always the same: why spend four years and hundreds of thousands of dollars when the internet gives you all the information in the world for free? It sounded compelling. Then college survived. MIT put entire courses online, YouTube exploded, Khan Academy appeared, and the best lectures on earth became available from your bedroom. Enrollment continued anyway.
The reason is simple. College was never really selling information. Information became free years ago. What college actually sold was credentialing, networks, structure, signaling, and a pathway into the professional world. The internet attacked knowledge. It never attacked capability.
AI might. That's what feels different this time.
A few years ago, if I wanted to build software, the path was established. Learn to code, take classes, read documentation, build projects, slowly accumulate the skills to make something useful. Today I can sit down with Claude and build a website, an application, or an automation without ever taking a computer science class. That's incredible. It's also strange. The internet let me learn anything. AI increasingly lets me do things without learning them first.
That distinction matters. If the internet made information free, AI is making capability cheap. And if capability becomes cheap, a lot of institutions have to answer a hard question: what exactly are you selling?
Universities aren't the only ones. The same question is surfacing across nearly every profession. Careers were historically built through apprenticeships. Not in the medieval sense, but economic ones. The first few years of most careers were spent doing low-value work. Analysts built models. Consultants built PowerPoints. Lawyers reviewed documents. Engineers updated CAD files. Junior software engineers fixed bugs. The work itself wasn't valuable. The learning was. Nobody paid young professionals because they were useful. They paid them because they were becoming useful. Every rung of the ladder taught you something needed for the next one.
So what happens when AI removes the bottom rungs? Everybody talks about the jobs AI will replace. The more interesting question is how people are supposed to learn the jobs that remain. If AI builds the first draft of the model, who learns financial analysis? If AI generates the code, who becomes the architect? If AI creates the CAD, who becomes the engineer? If AI reviews the documents, who becomes the lawyer?
For decades, expertise was built through repetition. The apprentice became the expert by doing the work. Now we're entering a world where the work may get done before the apprentice ever touches it. That isn't necessarily bad. In many ways it's amazing. But it creates a problem most institutions haven't acknowledged yet. The challenge isn't replacing experts. It's creating them.
This is where the college conversation gets interesting. The future university probably doesn't compete on information. That battle ended years ago. And increasingly it may not compete on capability either. If an AI system can help students write code, analyze data, design products, and build businesses, then teaching capability becomes a much less defensible business model.
The institutions that survive will focus on what AI can't easily provide. Networks. Trust. Judgment. Taste. Ambition. The ability to work with other people. The ability to decide what's worth building in the first place. Ironically, the future of education may become less technical and more human. Not because technical skills stop mattering, but because they're becoming easier to access. The scarce thing shifts.
That's what happened when information became free, and I think it's about to happen again. The internet made knowledge abundant. AI is making capability abundant. The next decade will be spent figuring out what becomes scarce. My guess? Not information. Not skills. People who know what to do with them.