Averaging Trap / Regression to the Mean

When decision-making or creative processes prioritize maximizing 'acceptability' or 'universal applicability,' outcomes get pulled toward the statistical median, losing distinctiveness and recognizability. AI systems trained on the average of countless 'good examples' rather than boundary-pushing innovations; organizations pursuing scale inadvertently flatten the distinctive edges of their culture.

Read the daily articles behind this idea on the Chinese edition.