The Next Revolution Will Be About Judgment

Every major shift in how work is organized inherits assumptions from the period before it.

Factories needed people in specific places at specific times because the machinery was in the building. Attendance was a practical requirement, not a management preference. Much of that thinking carried into knowledge work. The tools and the nature of value creation changed, but many organizations continued to treat visibility and productivity as closely related. Office buildings replaced factory floors, and presence remained the measure.

AI is another point where an inherited assumption stops fitting the environment.

Most discussion of AI concerns what it can do: write, summarize, analyze, and automate tasks that previously took hours. The more consequential question is what happens when the effort required to produce something falls while the value of good judgment stays the same.

If a person can complete in two hours what previously took eight, the organization has to decide what it is paying for: time, output, expertise, or decisions. The answer seems obvious until you examine how compensation, promotion, and performance systems are built. Most of them still measure some combination of time and output.

Knowledge workers have always been hired for a combination of execution and judgment. Execution was unavoidable because the tools demanded it. Reports had to be written, research gathered, information organized and communicated. AI removes part of that work. The skills that become more valuable are the ones associated with making sense of complexity: understanding tradeoffs, building alignment, recognizing patterns, and helping the organization make better decisions. That is judgment.

Judgment has a property that complicates the picture. It cannot be retrieved. It accumulates from working through problems, watching approaches fail, understanding why structures exist, making tradeoffs and living with the results, and carrying those lessons into the next decision. It forms in the difficult conversations, the discovery, the failed experiments, the disagreements, and the moment someone notices a pattern that others missed. It does not form in the summary written afterward.

So the tools that raise the value of judgment can also interrupt the process that produces it. Feed weeks of meeting recordings, research, design documents, and strategy discussions into a model and it will identify themes, generate recommendations, and produce something that reads as original. The output may be correct. The person holding it did not go through what produced it, and the next problem will differ enough that the artifact will not help.

This is not an argument against the tools. It is an argument for deciding deliberately which work to keep. An organization that automates the parts of a job where judgment is developed will eventually be short of the capability it now pays a premium for. Junior people will have fewer of the experiences that made senior people useful. The gap does not appear immediately. It appears years later as a shortage that no one can trace to a decision.

The industrial period optimized for labor. The knowledge period optimized for information. The current period is likely to optimize for judgment. If so, the organizations that adapt best will be the ones that learn to recognize judgment, reward it, and protect the conditions that produce it.