I was back in another meeting where AI appeared before the agenda did. Not because we were there to discuss it. It has simply become the slide every strategy deck seems to need, somewhere between growth and efficiency.
What I can't tell, sitting in those rooms, is whether we keep raising it because it matters that much or because everyone else is raising it. Either way, I am less convinced than I was that AI itself will be the advantage.
The models are improving fast, and they are becoming available to almost everyone. If that holds, the technology doesn't vanish as a differentiator overnight, but it probably matters less than we think. What becomes scarce instead is judgment. The model can return an answer in seconds, but knowing which problem deserved one, and having the discipline to ask whether a process should exist at all before automating it, is slower work and much harder to acquire.
For years software was expensive and hard to acquire. Today intelligence arrives through an API, and it's hard to imagine that staying a competitive advantage for long. Culture doesn't arrive through an API. Neither does taste, or instinct, or the readiness to admit a plan no longer makes sense.
That's why I suspect AI is becoming less of a technology challenge and more of a management one. The question isn't whether people have access to the tools. It's whether they are prepared to work differently because of them.
If intelligence becomes something almost anyone can buy, the difference is left to whoever decides best what to do with it. I'm not sure the people who run companies are ready to judge that.