The Uncomfortable Questions · An AI FAQ for leaders

Question #6 · Judgment

What’s one decision you’d never let AI make, and why that one?

The short answer

Decisions that carry moral weight need someone who carries the consequences.

Most leaders name something involving people: hiring, firing, telling someone hard news. That instinct is right. The real question is where you draw the line, and whether you draw it on purpose.

Norm E. Murray

The principle

Why it matters

Some decisions are judged not only by whether they were right, but by who stood behind them.

01

Accountability needs a face

When a decision changes someone’s livelihood, the person affected deserves to know a human weighed it and will stand behind it. A model can’t be held responsible.

02

Legitimacy comes from judgment

Even a correct decision can feel unfair if no one took responsibility for it. People accept hard outcomes more readily when they trust the person behind them.

03

Context lives outside the data

Decisions about people turn on things no system records well: a family situation, a team’s history, what was promised in a hallway conversation.

04

The line moves if nobody draws it

Without an explicit boundary, AI drifts from advising on these decisions to making them, one convenient step at a time.

The bottom line: AI can help prepare a hard decision. A person has to own it, and deliver it.

Your move

What to do about it

  1. Write the list.Name the decisions in your organization that always require a human owner.
  2. Define what AI may do around them.It can gather information or draft options, but a named person decides.
  3. Review the list each year.Capabilities change. Decide on purpose where the line sits, rather than letting it slide.
  4. Tell people.Employees and customers trust you more when they know which decisions a human makes.
  5. Deliver hard news in person.Whatever helped prepare it, the conversation itself belongs to a leader.