What if the biggest weakness in AI-powered innovation is that we are getting better at predicting human behavior while becoming worse at understanding what that behavior actually means?
In this solo episode of the Innovation Storytellers Show, I want to talk about something I believe is becoming increasingly important for innovation leaders: AI can identify patterns, surface signals, and predict what customers may do next, but prediction is not the same as understanding.
My background in anthropology has always shaped how I think about innovation. Anthropologist Clifford Geertz wrote about “thick description,” the idea that observing human behavior without understanding its context can lead us to misunderstand what we are seeing completely. A wink and a twitch may look almost identical, but they mean very different things.
AI faces a similar problem. A model might tell you that thousands of customers abandon their shopping carts at the same point. What it cannot reliably tell you is whether they are worried about money, overwhelmed by choice, consulting their family, reconsidering the purchase, or distracted by something happening in their lives.
The behavior is visible but the meaning often is not. This creates an important challenge for AI storytelling for business leaders and anyone responsible for storytelling for AI innovation. As organizations gain access to more customer data, they also need people who can listen, interpret, question assumptions, and understand the cultural context behind what the data appears to show.
In this episode, I share four ways innovation teams can interpret what AI cannot. I explain why AI should act as a scout rather than the storyteller, why customer stories should challenge data rather than confirm it, and why the strongest human-AI collaboration narrative combines machine intelligence with human judgment.
Your competitors can increasingly access the same models, similar data, and comparable technology. The advantage may come from what your people understand that the machines cannot fully explain.
If AI can tell us what customers are doing, are we spending enough time asking why?