the silent person in the room

friday i talked to a guest who has spent years teaching people how to make decisions. not just executives. everybody.

his line on the booking form is why i booked him. we keep investing in better leaders while we ignore the decision making of the people they lead.

i've sat in a lot of board rooms and a lot of team calls. i think he's right. and i think ai makes it worse before it makes it better.

he calls the skill decision literacy. critical thinking, working with uncertainty, knowing your biases, knowing your tools. and he says almost nobody trains for it. he sees students in india and the us show up with the same gaps. most of us decide the way we watched someone else decide, our first boss, the people around us, and we never question it.

then he said the thing i keep coming back to.

it's like having the silent person in the meeting. so maybe the silent person in the meeting would have a solution, but nobody would listen.

leadership training, he said, leaves that person out. you can make the person at the head of the table sharper and still lose the answer sitting three seats down, because that person doesn't have the context, or the confidence, or the language to put it on the table.

his fix is plain. make the decision process visible. get it out of people's heads and onto something everyone can see, the way you'd move sticky notes around a wall. when the stakeholders and the tradeoffs are out in the open, the quiet person can point at the board and say, that one's wrong.

the most common mistake he sees inside companies is the opposite. hiding things. especially during a merger or a big transformation. someone tells the team nothing will change. everyone knows that isn't true. so now every person makes their own calls from a distorted picture.

now add ai.

i asked him who owns the outcome when ai is part of the decision. his answer was simple. you do. a boss who takes a suggestion from someone on the team and then blames them when it goes wrong isn't off the hook. same with a model.

he described four ways people use ai. as a source of data. as a sparring partner. as a place to run scenarios. or as an agent you hand the job to. the further down that list you go, the more you need to understand what happens when it's wrong.

his best example was a group of experienced machine operators who didn't trust the ai recommendations they were handed. the company could have forced it on them. instead it combined what the operators knew with what the model saw, and left the final call with the people who had run those machines for years. human plus ai, human decides.

here's my take, as someone who builds with ai every day.

the model makes the loud people louder. whoever types the prompt gets an answer that sounds confident and complete. if your team's decision process is already a black box, ai just makes the box faster.

but open the process up and ai can do the opposite. it can hand the quiet person the research, the data, the counterargument they didn't have the standing to bring. it can level the room.

which one you get depends on whether you've built the habit of deciding in the open. the tool won't build that habit for you.

long term, the companies that win with ai won't be the ones with the best models. everyone will have the same models. they'll be the ones where the person closest to the problem can see the decision and say something.

more on using ai without handing it your judgment is in the ai book.

so think about your last big call... who was the silent person in the room, and did anyone ask them?