Systems thinking
The human element in systems
Large systems rarely fail because the technology stops working. More often, the breakdown begins in the human systems that interact with it.
Many program leaders have felt this moment: the architecture is sound, yet something in the system no longer feels stable.
That idea came back to me after two unexpected conversations. A financial executive commented on one of my posts and pointed to the structural dynamics behind the global financial recession. It sent me digging through old folders for something I hadn't opened in years—my master's thesis on the human element inside large institutional systems. A few days later, I had a fascinating conversation with an advocate for using AI for the good of humanity about how artificial intelligence is beginning to shape decision systems across industries.
Two very different discussions, finance and AI. Both pointed to the same systems question:
How do complex institutions stay stable when the environment around them is changing fast?
Three kinds of drift
When I revisited the thesis, one pattern stood out again. Long before anything breaks, three things start to drift:
- Mission incentives drift—the why.
- Institutional memory fades—the what.
- Communication pathways fragment—the how.
The technical architecture may still be sound, but the human dynamics inside the system lose their balance.
Homeostasis for programs
In biology, the body's ability to regulate itself under stress is called homeostasis. Organizations and large programs behave in remarkably similar ways. When the human networks inside them are healthy, the system adapts. When those networks weaken, instability can emerge even while the technology keeps working.
As AI becomes part of more decision systems, these human dynamics matter even more. A model can keep producing confident answers while the incentives, memory and communication around it quietly drift. The question for leaders isn't only "Is the model right?" It's "Is the system of people around the model still healthy enough to notice when it isn't?"
Have you watched a complex system begin to drift—not because the technology failed, but because something in the human dynamics changed? What was the first sign?
Is a model or AI output about to drive one of your decisions? Ten questions will tell you whether you're ready to own it.
Take the free Readiness Check