The Next Ten Years
I think the clearest way to describe the next ten years is this: there will be people who are fluent in how to create continually improving businesses, and there will be people who have not figured that out. The gap between those two groups is going to be significant. Not because the ones who have not figured it out are bad people or even lazy people. But because the leverage available to the fluent ones is going to compound in ways that make it very hard to close that gap over time.
The most specific version of this: if your team or your organization does not have someone who is extremely fluent in applied AI, you are going to struggle. Other teams competing in your space are going to have so much more leverage. They will iterate faster, serve customers better, and find efficiencies that let them do more with less. This is not hypothetical. It is already happening at the edges of most industries, and it is moving toward the center fast. The teams who are in this fine-tuned process of continually improving their business are going to pull away from the ones who are not.
What does fluency actually mean here? It is not about knowing every tool. It is about being in a continuous feedback loop with your business: using AI to identify what is working, what is not, what your customers actually need, and what is wasting your team’s energy. It is operational and strategic at the same time. The people who develop this fluency in the next few years are going to be extraordinarily valuable, whether they are running their own businesses or working inside larger ones.
I think there are two bifurcations worth naming here. The first is economic. AI is creating a K-shaped split where businesses either multiply their impact by orders of magnitude or they slowly cease to exist. There is very little middle ground forming. And the most expensive strategy right now is “wait and see.” Waiting feels prudent, but every month you wait, the teams that are already in motion compound further ahead. The urgency is real. But so is the opportunity, because we are still early enough that someone who commits today can catch the curve.
The second bifurcation is deeper, and I keep coming back to it. I believe there are two most important moments in human history: Jesus ascending to heaven and sending the Holy Spirit, and self-improving AI being invented. That framing might sound dramatic, but consider what self-improving AI actually means. Andrej Karpathy has been demonstrating auto-research systems that can fine-tune AI models without human help, where humans just describe objectives in plain English and the system iterates on itself. This is the worst AI is ever going to be. The pace of change only accelerates from here, because the technology is now capable of improving itself. Every week it gets better at getting better. If that does not reshape how you think about the next decade, I am not sure what will.
The spiritual underpinning of this, for me, is stewardship. Whatever assignment you have been given, whether it is a company you built or a role you occupy, you are being asked to steward it well. AI is one of the most powerful stewardship tools available right now. Using it poorly, or not using it at all, is a real cost.
Key Takeaway
The next decade will create a significant and compounding gap between teams that use AI to continually improve their business and those that do not.