What Gets Measured, AI Will Automate

In the age of AI, what gets measured gets automated. As models grow more powerful, any task that can be turned into data—from spreadsheet analysis to therapy sessions—is increasingly within reach of automation. The underlying playbook is clear: define the task, feed it data, attach rewards, and apply compute. As AI slashes the cost of measurement, even minor activities become economically viable to automate, expanding the reach of AI into nearly every industry. What remains defensible are tasks defined by ambiguity, creativity, or uncertainty—places where outcomes can’t be easily quantified or where human judgment still prevails. For leaders, the challenge is to manage both the measurable and the unmeasurable, investing not just in automation, but in the intangibles—taste, trust, vision, and adaptability—that AI can’t yet replicate.