A huge trend I am seeing in conversations with clients and prospects about leadership development, business acumen, and AI is the need for stronger critical thinking skills. People want their teams to ask
AI is adding another layer to these conversations. We can get information, analysis, and recommendations faster than ever with AI. But how do we make sure people are thinking through what they get back? And how do we help them use AI to expand their thinking without getting into the habit of accepting whatever it gives them?
It has me thinking about what we really mean when we say someone needs to develop their critical thinking. Is it a set of skills we can teach? A mindset we need to encourage? Or both?
I think it is both. We can teach people how to evaluate information and question assumptions. But they also have to be willing to do it, especially when the answer in front of them supports what they already believe.
Knowing how and being willing
For me, critical thinking comes down to taking a closer look before deciding what to believe, what to do, or what to prompt the AI. What is actually happening? What do we know? What are we assuming? What else could explain the situation?
These sound like basic questions. In the middle of a busy day, with a customer waiting and six other things that need attention, they are easy to skip.
Imagine a team that keeps missing deadlines. The first explanation is that people need to be more accountable. That may be true. But perhaps priorities keep changing. Perhaps three managers are assigning work to the same people. Perhaps the team is spending hours fixing problems that nobody has addressed.
If we accept the first explanation, we may put an accountability process around a workload problem. Now everyone has another meeting to attend, and the work is still late.
The skill is being able to examine those possibilities. The mindset is being willing to step back and say, “Maybe we have not figured this out yet.” That can be uncomfortable, especially if you were the person who offered the original explanation.
AI gives us an answer quickly. We still need to think about it.
I use AI, and I see tremendous value in it. It can help organize ideas, summarize information, analyze data, and get work moving much faster. There are plenty of tasks where I am happy to get those hours back.
But there is something very convincing about a well-written answer. It has a clear recommendation, supporting points, and a confident conclusion. It looks ready to go.
Sometimes it is. Sometimes it has built a very good argument around an assumption that nobody checked. Suppose customer support tickets are increasing. You ask AI to review the information, and it recommends more customer training because most tickets are labeled “training support.”
Seems reasonable.
But does that label tell us what caused the problem? Or does it tell us what the support team did after the customer called? Have those customers already been trained? Is something about the product or installation creating the confusion?
Those questions could lead to a very different decision. AI helped us get through the information faster. We still have to determine whether the recommendation addresses the problem.
We also need to question the answers we like
This may be the hardest part. If I think pricing is the reason sales are down, and AI gives me five reasons our pricing may be hurting sales, I now have a nicely organized case for what I already believed.
But what about service? Availability? A change in the customer’s business? A change in how are customers are consuming simulations? Are we losing across the entire market or in one segment?
A useful way to work with AI is to ask it to challenge your interpretation. “What am I assuming?” “What else could explain this?” “What evidence would suggest that my recommendation is wrong?”
Then you have to examine those responses, too. Asking AI to challenge itself does not automatically make the next answer correct. It gives you more possibilities to investigate. That is where the mindset comes in. You have to be interested in finding a better answer, even when it creates more work or means changing your position.
How do we help people develop this?
I would start with the conversations already happening at work.
When someone brings a recommendation, ask them to walk through how they got there. What information mattered most? What other explanation did they consider? What are they still unsure about?
We also need to pay attention to how leaders respond. If we ask people to challenge assumptions but get frustrated when they question ours, they will figure that out pretty quickly.
The same goes for changing a recommendation. If someone finds new information and adjusts their thinking, that should be part of making a good decision. People should not feel they have to defend their first answer forever.
Practice matters as well. Give people a realistic situation where the information is incomplete, colleagues have different views, and a decision needs to be made. Let them ask questions, choose what to investigate, and explain their recommendation. Then introduce something new and see how they respond.
This is a good fit for simulation-based learning. Participants can experience how an assumption affects a decision and what happens when they fail to ask an important question. The feedback can focus on their thinking: what they noticed, what they accepted too quickly, and what they could have explored.
Over time, those are the habits we want people to bring back to work. AI will continue to help us do more, faster. The opportunity is to use some of that time to look more closely, consider another perspective, and ask the question we might otherwise have skipped.
Before we act on the next polished recommendation, it is worth asking: “Do I understand why this makes sense?” And being willing to take another look if the answer is no.