Not long ago, if we wanted to do something, we had to learn how to do it first. Writing code, designing, analyzing data — all of these took time and practice. Today, we can simply tell AI, “This is what I want,” and let it do much of the work.
AI keeps getting better at things we once thought only humans could do. And with every new capability, it’s hard not to wonder: is there anything AI can’t do?
In a way, AI makes us feel like we can do almost anything, even things we barely understand. We can build something without knowing how it works, or get answers without understanding how they were reached. We can do almost anything, yet understand so little.
But maybe that’s not the right question. AI will keep improving, and the list of things it can do will only grow. Perhaps a better question is: what should we actually leave to AI?
AI can write a message of condolence, but do we really want words meant to comfort someone who’s grieving to come from a machine? AI can summarize hundreds of pages in seconds, but it can also be confidently wrong. And when something goes wrong, AI isn’t the one dealing with the consequences. We are.
That’s where the line becomes clearer. Not in what AI is capable of, but in who takes responsibility. AI can help us make decisions and get things done, but it can’t take responsibility for the outcomes.
This doesn’t mean we should stop using AI. It can save us time, make our work easier, and help us focus on what matters. But being able to automate something doesn’t always mean we should.
So, can AI do anything? Maybe one day.
But perhaps the more important question is: what do we still want to do ourselves?