A few days ago I spoke to first-year students on the opening day of their year. Eighteen, most of them. Some had come for computer science, some for the humanities, some for management or commerce or the pure sciences. Bright, a little restless, and all of them carrying the same question without quite saying it: now that AI has shown up, what happens to my career?
I want to write down what I told them, because I don't think it's a question only eighteen-year-olds are asking.
Start somewhere basic. Why do jobs even exist? Strip everything away and a job is just a problem somebody is paid to solve. People need food, health, a roof, education, safety, someone to talk to. Those needs don't go anywhere, and every one of them is somebody's work. What's easy to miss is that the needs stay put while the way we meet them gets rewritten every couple of generations. Farming gave us a food surplus. The steam engine gave us factories. Electricity, mass production. Computers, the information age. Now it's AI's turn in the queue. Every one of those shifts scared the people living through it, and every one of them ended up making more work than it destroyed. Just different work. Technology has never really deleted human work for good. It keeps moving it around.
I'm not going to pretend this round feels comfortable, because it doesn't. Last year, at college graduations across America, students actually booed the speakers who praised AI. One called it "the next Industrial Revolution" and got booed. A former Google CEO talked about the people building AI, booed. A music executive said AI was rewriting how records are made, booed again. The videos went everywhere. One news channel said the word "AI" had become "wildly unpopular." So if you've got a knot in your stomach about all this, relax, you're in large company. But booing the wave has never once stopped the wave. You can stand on the beach and shout at the tide all you like. It comes in anyway. The only useful thing to do is learn to swim in it.
And that begins with a small change in the question you ask yourself. Most people are quietly running this one: will AI take my job? Drop it. Not because the worry is silly, but because it puts you in the back seat and hands all the power to the machine. Ask the other one instead, the one that puts your hands back on the wheel: how do I create value in a world that now has AI in it?
I'm suspicious of hype in both directions, so let me reach for some evidence. The World Economic Forum looked across a thousand of the world's biggest employers. Their estimate for 2030 is roughly 92 million roles displaced, but around 170 million new ones created. Net, about 78 million more jobs than before. The Indian picture is sharper: by 2027 the country is expected to have over 2.3 million open AI-related roles and only about 1.2 million people skilled enough to fill them. That's a gap of over a million jobs, and somebody has to stand in it. Why not you? And the roles growing fastest aren't the purely technical ones, McKinsey finds. They're the hybrids, the ones mixing real depth in a field with comfort around AI.
Here's an old story people forget. When ATMs arrived, everyone assumed bank tellers were finished. They weren't. Running a branch got cheaper, banks opened more of them, and the number of tellers actually went up for years even as the machines spread. The job didn't vanish. It shifted, from counting cash to looking after customers. That's roughly the shape of what's coming for the rest of us, because AI doesn't replace people. It replaces tasks. Think of a teacher. AI can draft lesson material now, churn out quizzes, even tailor practice to each student. But it can't mentor a teenager who's struggling, or notice a kid has had a rotten week, or make someone fall in love with a subject. Most of your careers will live in that middle ground, where people and AI work together.
This is the part I most wanted the non-coders in that hall to hear, so let me be plain. If you're in the humanities or social sciences, AI is quietly opening whole new fields for you: AI ethics and policy, governance, the digital humanities, communication, the design of how people learn. AI throws up the hardest questions we've ever had about meaning, fairness, language, culture and law, and those are precisely your questions. If you're in management or commerce, someone has to decide what to build, lead the project, set the strategy, weigh the risk and walk real people through real change. A model doesn't do that. People do. The little formula I'd ask you to keep is this: your field, plus AI literacy, is your edge.
And even for the coders, I'd push back on the doom. Writing in The Atlantic, Lila Shroff argues the "computer science is dead" story has been badly oversold. Fresh CS graduates still have low underemployment and still out-earn their peers. As more of the economy turns into software, we'll need more people who understand systems deeply, not fewer. AI is eating the line-by-line typing of code. It isn't eating the thinking. And that's really the same shift hitting everyone, in every subject. Away from doing the routine, towards understanding the system and using your judgement.
One caution before you dive in, the engineer in me has to say it. Use the thing like a scientist. It hallucinates, and will tell you something completely false with a straight face. It carries whatever bias was in its training data. It gets shaky on anything niche, and it'll happily swallow private information you should never have given it. One rule covers all of that: treat every output as a draft to be questioned, never an answer to be trusted. The people who get the most out of AI aren't the ones who believe it. They're the ones who know how to check it.
So what do you actually do? If I could hand a fresher one habit, and really anyone reading this, it would be to learn, unlearn and relearn, and keep doing that for the rest of your life. Any single skill goes stale faster than it used to. The knack for picking up the next one doesn't. Build the human muscles around it too, the thinking and communicating and curiosity and the sense of right and wrong, because AI is wonderful at chewing through information, but it's still people who supply the meaning and the trust, and those only get more valuable as the machines get cheaper.
And please don't spend these years just chasing a job. Try to become valuable. A job can be handed to you and taken back. The ability to solve real problems for real people can't. Jobs will change, tools will change, whole industries will rise and quietly die over the length of your career. But the people who keep learning, keep adapting and keep making themselves useful tend to stay relevant in every era I can think of.
The future doesn't belong to AI on its own. It doesn't belong to us on our own either. It belongs to the people who learn to work intelligently with it.
There's real gold here. There's also a lot of noise. 🙂
Navigate wisely.
Comments