Until last year, I thought my accomplishments as a software developer peaked in third grade, writing BASIC programs on our IBM PC. As an adult the translation from idea to functional code has been a steep hill to climb. Yet today, I find myself building things that actually work. I'm not a good programmer by any stretch, but I've become good enough to bring my ideas to life.
This transformation wasn't the result of thousands of hours of study and practice. It happened because of AI tools that elevated my capabilities just enough to cross the threshold from "can't do it" to "can make it work."
This got me thinking about expertise in this changing world.
An MIT Sloan study examined how AI coding tools affect software developers. The researchers found these tools increased productivity across the board, but junior developers and those with less experience saw much greater improvements than senior developers. This tracks with my own experience.
On the surface, this might suggest AI primarily benefits novices in a specific domain. But I think there's a more interesting story here if we zoom out and consider the full spectrum of skills that make up an individual's capabilities.
None of us are experts at everything. Even the most talented individuals typically excel in just a few areas while being average or below average in many others. As a species we've found specialization to be tremendously beneficial.
But there are degrees of specialization, and it turns out there are many situations where good enough is just fine.
As a recovering perfectionist, the 2009 Wired article "The Good Enough Revolution" made a big impression on me. Robert Capps explored how many markets were being transformed by products that traded high fidelity and extensive features for convenience, accessibility, and lower costs. The article pointed to examples like the Flip video camera, MP3s, and cloud computing - all technologies that weren't technically "the best" but were accessible enough to change consumer behavior:
We now favor flexibility over high fidelity, convenience over features, quick and dirty over slow and polished. Having it here and now is more important than having it perfect. These changes run so deep and wide, they're actually altering what we mean when we describe a product as "high-quality."
AI is taking this "good enough" concept to our skills and capabilities. It doesn't make us experts at everything, but it can make us functional across a much wider range of activities than was previously possible.
The MIT study showed AI doesn't offer much productivity gain to expert coders. But what if that expert coder is also creating marketing copy, basic graphics, analyzing user data, drafting terms of service, or composing investor pitches? Now, the value proposition changes dramatically. AI might not significantly improve their coding, but it could elevate their performance across all these adjacent tasks to "good enough."
For now, this is where the real transformation is happening - not in making experts marginally better at their specialties, but in raising the floor on everything else they can do. Those who can operate competently across multiple domains and synthesize between them can have more impact than ever.
The future might not belong to those who can do one thing perfectly, nor to those who can do everything adequately, but to those who understand which skills matter most in their domain and thoughtfully use AI to extend their capabilities everywhere else. Just as AI helped transform me from someone who couldn't code into someone who can build functional software, it's expanding the surface area where each of us can be effective.