Longer-form thinking on the human capabilities that make data and AI actually work — and the organizational systems that determine whether they get used well.
Most organizations have invested heavily in data infrastructure, yet their decisions remain largely unchanged. The reason has less to do with technology than with how decisions actually get made — and the cognitive habits, organizational structures, and incentive systems that surround them.
Read the essayMost organizations have invested heavily in data infrastructure, yet their decisions remain largely unchanged. The reason has less to do with technology than with how decisions actually get made.
Generative AI is shifting what gets automated. Not the work, but the thinking. And the organizations that thrive will be the ones that notice the difference before the muscle goes soft.
Data literacy is not the ability to read a chart. It is the cognitive habit of distinguishing signal from noise, evidence from anecdote, and confidence from certainty.
For two decades, organizations have responded to decision quality concerns by producing more information. The trouble is that the gap was never about information. It was always about what people do with it.
Some organizations notice change earlier and adapt faster than others. The difference is rarely about resources or talent. It is about how the organization itself thinks as a system.
Eighteen months after launch, most enterprise AI programs have changed surprisingly little about how the business actually operates. The pattern is consistent enough to qualify as a law.
AI can act as a tool, an advisor, or a decision-maker. Each mode shifts the burden on human judgment differently. And most organizations have not yet thought carefully about which mode they actually want.
Across dozens of organizations, the same architectural patterns keep failing in the same ways. Naming them is the first step toward designing decision systems that actually work.