The traditional approach we’ve long followed is: learn first, then practice. When problems arise in practice, we return to learning. Theory is the starting point; action is the validation. This model made perfect sense in an era of knowledge scarcity and high trial-and-error costs.
But with the emergence of AI, this path is being rewritten.
Today, a more effective approach is often: practice first, learn, then practice again.
You don’t have to wait until you’ve “learned” something to begin. With AI, you can build, ask, and iterate simultaneously. Practice itself becomes the gateway to learning.
As the cost of acquiring knowledge continues to decline, what's truly scarce is no longer information—but action and feedback. Rather than spending vast amounts of time preparing, it’s better to start early, generate learning needs from real problems, and then quickly close cognitive gaps with AI.
Learning is no longer a linear accumulation, but an iterative loop.
In the AI era, knowing how to use tools, having the courage to act, and doing enough of it—these are becoming more important than “how much you’ve learned.”