One of the readers of my recent anniversary article posed an interesting observation. Uber was not possible until smartphones became common. In hindsight, Uber seems obvious. But in 2005 it would have sounded ridiculous. Millions of strangers getting into the cars of other strangers, coordinated through a handheld computer connected to satellites and payment systems? The idea depended on several technologies maturing at the same time. Once they did, an entirely new business model emerged. Without GPS, mobile internet, digital payments, and millions of connected users carrying computers in their pockets, the business simply could not exist. The same could be said for Airbnb, food delivery apps, mobile banking, and dozens of other services we now take for granted. Which raises an interesting question. If…

Technology should serve people.
People should not be forced to serve technology.

Technology often changes products first. Then it changes systems. Eventually it changes expectations.

The question is whether our institutional systems can evolve fast enough to manage large-scale infrastructure development in a world where technological change is accelerating much faster than regulatory adaptation.

If AI follows the same pattern, what do we do differently? In the last article, I ended with a question. If major changes tend to unfold gradually—and if we usually don’t recognize their full impact until they reach us—then what does that mean for artificial intelligence? More specifically: If AI is following a familiar pattern, is there anything we can do differently this time? We’ve Been Here Before History suggests a consistent response. We don’t ignore change because we’re unaware of it. In most cases, the signs are visible. New technologies emerge, early adopters experiment, and initial use cases begin to take shape. What we tend to do instead is wait. We wait until the change becomes relevant to us personally. Until it affects our…

We’ve seen the pattern before—just not always in time Large changes rarely feel important while they are happening. Not because we don’t see them, but because, at least at first, they tend to happen to someone else. A farmer loses his land. A small operator sells off equipment that no longer makes sense. A business across town closes. A new way of doing things appears, and it works—but not in a way that seems immediately relevant. All of these are real changes. They’re just not our changes. However, the question is are we not experiencing a big change in progress? As long as changes from AI stay in controlled spaces and remain that way, it’s easy to assume that whatever is happening is limited, temporary,…

On AI, Security, and the Problems We Didn’t Look For A recent article that showed up in my email feed caught my attention. It said the latest release of Claud Mythos has capabilities that are too dangerous to mke broadly available until our most important software is in a much stronger state. AI systems—specifically tools like Claude—are being used to analyze existing code and uncover security vulnerabilities at a scale that wasn’t practical before. That, by itself, is interesting. But there’s a second layer that may be more important. In simple terms, once the barn door is open, it tends to stay open. Once that capability is shown to exist by Claude, it doesn’t stay contained. Other developers will build similar tools. Other organizations will…