How one mistaken judgment reshaped the way I think about technology, the future, and myself

An era does not slow down just because you find it uninteresting. It simply leaves you standing on the shore.

In my sophomore year, I was still immersed in mathematics courses: mathematical analysis, linear algebra, ordinary differential equations, probability. I loved the purity of logical deduction and the excitement of seeing a proof finally click. At the time, that was the substance of my daily intellectual life. I often imagined that one day I might use mathematics to push technology forward. But my understanding of technology itself was still shallow, limited to a few distant terms from the news. Artificial intelligence felt like a storm happening somewhere far away, and I was just an outsider watching from the edge. My memory of AI was still tied to the amazement around AlphaGo defeating Lee Sedol, and occasional headlines about AlphaFold. Since I did not understand them, I instinctively ignored their importance. I did not realize that the tide of the era had already begun to rise, and I was standing right beside it.

By chance, I entered an interdisciplinary "AI + X" micro-program jointly launched by top universities and industry. The courses covered GANs, Transformers, and even QKV attention and multi-head structure, all of which were already frontier topics at the time. But as a mathematics student, what kept catching my attention was not what these ideas could do, but whether the mathematics behind them felt deep enough. What I saw were relatively simple linear algebra operations and straightforward gradient calculations, so I came to an easy conclusion: this was not very complicated, and probably not the kind of thing that could lead to truly important work. I failed to see that the power of AI did not come from mathematical difficulty. It came from the scale of problems it could solve and the leverage it could create. After the course, I could already build small neural networks for classification and regression tasks, yet I remained convinced that AI was merely a tool for solving minor problems, not something worth committing myself to for the long term. So I turned away and chose to focus on computational mathematics, which felt to me like the truly advanced and rigorous field. What I underestimated was not just a technology, but my own bias toward the unfamiliar. I was using the kind of complexity I already understood as my standard for value, while ignoring the possibility of growth.

The years that followed brought a much faster and more complete AI explosion than I had imagined. GPT-3 began attracting serious attention, GPT-4 and DeepSeek made global impact, Stable Diffusion brought image generation into ordinary hands, robotics accelerated rapidly, and AI agents began reshaping the way many industries work. The technology I had once dismissed was now helping redefine the world. Watching people who had learned alongside me move quickly with the AI wave, I finally understood how narrow my judgment had been.

Looking back, I now understand that what I lost was not a technical skill, but the ability to recognize future potential. I wrongly believed that the value of a technology was proportional to the sophistication of its mathematics, so I rejected models that seemed too simple to me. I judged long-term development from a crude early form and ignored the shape of technological growth curves. I resisted innovation outside the framework I already knew, as if only what matched my existing notion of complexity deserved attention. These biases made me miss a crucial window in AI's development. The real dividing line between people is rarely current technical difficulty. It is whether they can see potential, understand growth, and dare to move toward it.

I cannot go back to that year. But I can change the way I look at the future. I have learned to ask different questions: Is this technology solving a real social need? Does it have leverage for large-scale growth? Could its future form become much larger than what it looks like today? Instead of dismissing something as "too simple," I now try to understand what it may become. Many forces that eventually reshape the world do not look impressive at first glance. They grow in ways we do not yet know how to read. The only thing we can do is give them room to grow, and train ourselves to recognize that growth.

I once missed an era because I said, "This is too simple." Now I understand that what truly separates people is not momentary cleverness, but the ability to sense potential and give the future time and space to mature. If you are facing a new trend now, one that seems rough, immature, or inelegant, do not rush to turn away. It may be the beginning of the next wave. The ticket to the next era might already be in front of you. The only question is whether you will step in, or once again say "too simple" and walk away.

贡献者: Junyuan He