AI and the Psychology of Judgement
AI changes the conditions for honest self-evaluation — and makes deliberate judgement more important than ever.
One of the more interesting things I've noticed about working with AI isn't what it does to productivity.
It's what it does to our judgement.
We are not objective evaluators of our own work. This isn't a character flaw — it's well-documented psychology. We become attached to ideas because they feel like ours. We confuse familiarity with quality. We look for evidence that confirms our initial excitement and discount evidence that doesn't. The investment we put into making something distorts how clearly we can see it.
This was always true. What's changed is the environment in which it plays out.
Historically, creating something professionally meant living with it for a while. You'd draft, revise, abandon, restart. The process was slow and often frustrating, but it had an important side effect. It gave you repeated chances to encounter your own weaknesses. A phrase that seemed brilliant on Tuesday felt clumsy by Thursday. A structure that looked elegant in outline fell apart in execution. Time created distance. Distance created something closer to honest appraisal.
AI compresses that cycle.
You can move from idea to polished output in minutes, and the work genuinely feels like yours, because in the ways that matter, it is. Decisions were made. Judgements were exercised. Something exists that didn't exist before. But the pauses that once came naturally through effort no longer arrive automatically. Reflection becomes something you have to choose, rather than something the process imposes on you.
Because consider what happens next.
You share the work. And the person reading it encounters something completely different from what you made. They have no memory of the six directions you didn't take. No context for why a particular choice mattered. They open a document, or a headline, or thirty seconds of video, with no investment in your process and only a reaction to your output.
They experience it the way a stranger does. Which, almost certainly, is more accurately than you do.
That gap between how a creator sees their work and how an audience actually receives it has always existed. But it grows when the effort that used to anchor self-assessment disappears. When there's no long night of drafting to remind you that good work is hard, it becomes easier to assume that fast work is good work.
What makes this particularly worth sitting with is that AI doesn't resolve this tension. It sharpens it in both directions.
Used carelessly, it accelerates existing bias. It helps you produce more work while spending less time questioning whether the work is actually good. The excitement of fast creation masks the absence of genuine evaluation.
Used deliberately, it can do the opposite. It can challenge your assumptions before you've committed to them. Surface the counterargument you hadn't considered. Play the sceptical reader rather than the enthusiastic collaborator. The same tool that compresses reflection can, if you ask it to, replace it.
The technology doesn't determine which of those paths you take.
You do.
So I don't think this is an argument against AI. If anything, it's an argument for understanding ourselves more clearly.
Perhaps one of the most important skills in this period isn't learning how to create faster. It's learning how to evaluate more deliberately, to manufacture the distance that used to arrive on its own, to invite critique when praise is easier, to slow down at exactly the point where everything is telling you to share.
And to ask yourself, honestly, before you do:
If this wasn't mine, would I still think it was good?
Because AI may have changed the speed of creation.
But it hasn't changed the psychology of judgement.