What is critical thinking?
Critical thinking is the disciplined practice of deciding what to believe or do by examining the reasoning and evidence behind a claim — rather than accepting it at face value.
That's the short answer. The longer one is more useful, because critical thinking isn't a single skill so much as a handful of mental moves you can learn to make on purpose. The opposite of critical thinking isn't stupidity; it's automatic thinking — taking the first plausible answer, the confident voice, or the conclusion you already favoured, and stopping there.
The core moves
Strip away the jargon and most accounts of critical thinking describe the same five habits:
Clarify the question. Be precise about what's actually being asked or claimed. A surprising amount of muddled thinking is really a muddled question — vague terms, two arguments tangled into one, or a problem nobody has defined.
Surface the assumptions. Every claim rests on things taken for granted. Naming them — "this only follows if X is true" — is often where a weak argument quietly falls apart.
Weigh the evidence. Ask what would actually support or undermine the claim, how good that evidence is, and where it came from. A confident assertion and a verified fact are not the same thing, however similar they sound.
Consider the alternatives. Hold more than one explanation in mind before choosing. The first answer that fits is rarely the only one that fits, and treating it as such is how good reasoners still reach bad conclusions.
Check your own reasoning. The hardest move: turning the scrutiny on yourself. Am I believing this because it's well-supported, or because it's convenient, familiar, or flattering?
Critical thinking is less about being clever and more about refusing to let "it sounds right" be the end of the matter.
What it isn't
Two confusions are worth clearing up. First, critical thinking is not the same as intelligence. Intelligence is horsepower; critical thinking is keeping your hands on the wheel. Clever people reason badly all the time — usually by deploying that horsepower to defend a position rather than test it.
Second, it is not the same as cynicism. Reflexively doubting everything is just as automatic as believing everything; it simply fails in the other direction. The point isn't to trust nothing — it's to apportion your trust to the strength of the reasoning and evidence. Sometimes that means believing a well-supported claim wholeheartedly.
Why it matters more in the age of AI
For most of history the friction of finding an answer did some of this work for you. You had to seek out a source, read it, weigh it. AI removes that friction entirely. You now get an articulate, confident, comprehensive-sounding answer to almost any question in seconds — and that answer may be wrong, may be shaped by how you asked, and is often tuned to agree with you.
So the economics of thinking have flipped. The effort needed to accept an answer uncritically has fallen to nearly zero, while the volume of plausible-but-unchecked material flowing past you has exploded. In that environment the five moves above stop being an academic nicety and become a practical defence — the difference between using a powerful tool and being quietly steered by it.
That's the specific problem this site is built around. The classic models of critical thinking still hold, but they predate fluent machines. We use a framework called RADAR to name the five places judgement tends to slip at the human–AI interface — calibrated trust, independent thinking, spotting fabrication, noticing how your framing shapes the answer, and resisting flattery — and a short reading that shows you where yours stands today.
How to get better at it
Critical thinking improves the way any habit does: by being made deliberate until it becomes automatic in the better direction. (Unfamiliar with a term used here? The glossary defines the key ones.) A few starting points. Before accepting an answer, ask "what would have to be true for this to be wrong?" Form your own view before you look one up, so the first answer you see informs your judgement instead of replacing it. When something confirms what you hoped, treat that as the cue to look harder, not to stop. And get specific about evidence — trace a striking figure or quote to where it actually came from rather than to where it was repeated.
See where you stand. The RADAR reading turns these habits into something measurable — a seven-minute, situational assessment of how well you keep your own judgement in the loop while working with AI, with one concrete habit to work on at the end.