The Value of Human Judgment in the Age of AI

Two days ago EY announced they will give $100M in bonuses for employees who show adaptability, innovation, and JUDGMENT.

This immediately made me think of a book I read in 2018. It said machine generated predictions were becoming cheap while human judgment was becoming much more valuable.

The book showed a clear role for both machine (prediction) and human (judgment) in the future - especially for enterprise decision-making.

"Prediction facilitates judgment by reducing uncertainty, while judgment assigns value. In economists parlance, judgment is the skill used to determine a payoff, utility, reward, or profit. The most significant implication of prediction machines is they increase the value of judgment."

After reading the book in 2018, I felt confident that human judgment would remain king. Then came LLMs and reasoning models. I've been thinking a lot about how these technologies change the equation as agents are now deciding and acting.

While a lot has changed in AI technology from 2018 to today, the clear signal from EY about the value of human judgment was refreshing to me. We often hear terms like "AI-first". It can sound like humans are just serving the machines.

We all know that human judgment is essential in the age of AI but for a company to come out and say it is good in my humble opinion.

Nice work EY.

Now I am inspired to go back and revisit the updated and expanded edition of this book. The AI era is here.

Figuring out where to use humans and where to use AI in the enterprise is an ongoing discussion.

The updated edition is from 2022 as is their newer book Power and Prediction. I'm going to revisit both and see how they held up.

Guidance from 2022 should be viewed from the lens of the enormous AI advancements in the last 4 years. Yet, frameworks can be really helpful when looking at these kinds of hard questions.