For most of my life, I believed that success came from learning.
Learn a new programming language. Learn a new framework. Learn a new methodology. Learn a new design technique.
In technology, there is always something new to learn, and I loved it. Curiosity has served me well throughout my career.
But over time I noticed something surprising.
The biggest breakthroughs in my life didn’t happen when I learned something new. They happened when I let go of something old.
I had to unlearn.
The first major example came when I moved from procedural programming to object-oriented programming in the early 1990s. At the time I was a successful C programmer. I understood procedural decomposition. I understood control flow. I understood debugging.
Then object-oriented programming arrived and challenged many of my assumptions.
Tracing execution through a program had been my security blanket. That’s how I understood systems. Object-oriented design required me to think differently. I had to learn to trust abstractions. I had to think in terms of responsibilities, collaborations, and behavior rather than simply control flow.
To move forward, I had to let go of something that had worked well for me.
Years later, I experienced a similar shift when I encountered design patterns through the work of Al Shalloway and Scott Bain. At first, patterns looked like reusable solutions. Over time I realized they were something much deeper.
Patterns taught me how to see.
They helped me recognize forces, variation points, relationships, and structures that had previously been invisible to me. Once again, growth required me to reconsider assumptions I didn’t even realize I was carrying.
The pattern repeated itself. Again and again. Now AI is asking us to do it on an even larger scale.
The Hidden Cost of Expertise
One of the great benefits of expertise is that it helps us make sense of the world.
One of the dangers of expertise is that it can convince us we already understand it.
The things we know create filters through which we interpret new information. Most of the time those filters are useful. They help us move quickly and make decisions.
Sometimes they become limitations.
I noticed this when I was in college. The things I already knew about a subject often made it harder to learn something new about it. My existing mental models would automatically reinterpret new ideas in terms of what I already believed.
The problem wasn’t a lack of intelligence. The problem was attachment. I was attached to explanations that had worked in the past.
Scientists encounter this challenge constantly. Good science requires the willingness to ask:
“What if my current explanation is incomplete?”
“What else could this mean?”
“What assumptions am I making?”
Those questions are uncomfortable because they threaten certainty. They are also the source of discovery.
Beginner’s Mind
There is a concept in Zen Buddhism called beginner’s mind. The idea is simple but profound.
The beginner sees possibilities. The expert sees conclusions.
The beginner asks questions. The expert often assumes answers.
This doesn’t mean expertise is bad. Expertise is valuable. The goal is not to abandon knowledge.
The goal is to hold knowledge lightly enough that we remain capable of learning. The most impressive people I’ve met throughout my career all seem to share this quality.
They remain curious. They continue asking questions. They are willing to reconsider deeply held assumptions. They maintain a beginner’s mind even after decades of experience.
AI and the Capability Shift
Many people see AI primarily as a technology revolution.
I think it is also a capability revolution.
For decades, organizations rewarded the accumulation of knowledge.
Learn more. Know more. Become the expert.
That made perfect sense when knowledge was scarce.
Today knowledge is increasingly abundant.
AI can explain concepts, summarize information, generate ideas, and answer questions in seconds.
This doesn’t eliminate the need for expertise. But it does change where value comes from.
The scarce resource is no longer information.
The scarce resource is capability. Judgment. Adaptability. Creativity. Communication. Learning. Experimentation.
The ability to make sense of complexity. The ability to collaborate effectively with both humans and AI.
These capabilities cannot simply be downloaded. They must be developed.
## Why Some People Thrive and Others Struggle
I’ve been thinking a lot about why some people are energized by AI while others feel threatened.
Part of the answer may be identity.
If my value comes primarily from what I know, AI can feel threatening. If my value comes from my ability to learn, adapt, and grow, AI can feel empowering.
The difference isn’t the technology. The difference is our relationship to learning.
People who embrace continual growth have often been practicing unlearning for years.
People who have built their identity around certainty may find this transition much more difficult.
The Most Important Question
The question facing us is not: “How do we keep up with AI?”
The more important question is: “How do we remain capable of learning?”
Because AI will continue changing. Technology will continue changing. Markets will continue changing. Organizations will continue changing.
The ability to reconsider, adapt, and grow may be the most durable capability we can develop.
I’ve spent much of my career teaching software development, design patterns, Agile practices, and software craftsmanship. Looking back, I realize those subjects were never really about technology. They were about learning how to see.
AI is teaching us the same lesson. The challenge is not simply learning something new. The challenge is becoming the kind of person who can continually rediscover the world.
That requires knowledge. It requires experience. And sometimes, it requires the courage to let go of both.
Because every stage of growth asks us to release something before something larger can emerge.
Maybe learning isn’t the ultimate superpower.
Maybe unlearning is.
