Critical Thinking Is Your Edge in the Age of AI

Imagine meeting a stranger and, within minutes, telling them your date of birth, exactly where you live and the details of a conversation you would never want made public. They listen patiently. They seem knowledgeable. They offer to help. Would that be enough for you to trust them? If the answer is no, what happens to our critical thinking when the stranger becomes a chat window?

The comparison has limits: AI services have different privacy policies, settings and protections. But the pause before sharing matters. Do we understand where that information is going? Is it ours to disclose? Convenience can make those questions feel unnecessary precisely when we should be asking them.

I remember hearing a UNICEF representative speak about critical thinking as artificial intelligence develops. It resonated with a question I hear from business leaders: should employees put company information into AI applications? The answer requires judgement about the information, the tool and the people affected. Having somewhere to paste a document does not settle whether we should.

When a convincing answer feels like understanding

The attraction is easy to understand. In Microsoft’s 2026 Work Trend Index, 58% of respondents said AI helped them produce work they could not have produced a year earlier. The survey covered 20,000 knowledge workers using AI at work across ten countries. These are people’s reported experiences, but they help explain why the technology is becoming part of everyday decisions.

In the same survey, 46% identified critical thinking as a human skill becoming more important as AI takes on more work. Recognising its importance, of course, is different from practising it when a deadline is approaching and an answer already looks finished.

A fluent response can make an assumption difficult to spot. Imagine a founder asking whether to launch a premium service. AI produces a persuasive case: a growing market, an underserved audience, an attractive opportunity. Yet the decision may depend on something that remains untested: will enough people actually pay that price?

That is where I would start. What would change my mind? A conversation with prospective customers or a small trial could reveal more than another polished forecast. If every question we ask is designed to confirm the launch, we are using research to reassure ourselves.

The judgement we still have to make

Before asking AI for advice, try writing down what you think and why. Then compare the response with your reasoning. Has it introduced evidence, challenged an assumption or simply made your existing opinion sound more impressive? Open the sources behind the claims that matter. Check their dates, who they describe and whether they support the conclusion being drawn.

For leaders, there is another test: can someone challenge the answer after you have endorsed it? Asking a team to think critically means making room for the colleague who notices a weak assumption and slows an exciting plan down.

I want the time AI saves to give us more room for that kind of thought. Before putting your name to its recommendation, close the chat and explain the decision in your own words. If you cannot, what exactly are you agreeing to?