Interactive Learning
Test your knowledge across 3 games. You have 3 โค๏ธ to complete the challenge.
The Critic
You've made something. Now you want to know if it's any good.
That's the Critic. Not the role that makes the thing for you, and not the one that teaches you. The one that looks at what you've already done and tells you the truth about it.
Reach for the Critic when the work already exists
The essay is written. The email is drafted. The design is done, the code runs, the photo of the thing you built is sitting on your phone. Now you want eyes on it.
Feedback. A second opinion. A check before you hit send or hand it in. Anytime there's something finished in front of you and the question is "is this actually good, and what's wrong with it," you want the Critic.
The Critic's enemy is politeness
Here's the catch that ruins most feedback.
Ask the AI "what do you think of my essay?" and it'll tell you it's wonderful. A strong opening, a clear voice, just a few tiny suggestions. It's built to encourage you. By default, it's a cheerleader, not a critic.
A cheerleader feels great and changes nothing.
So the single most important thing you do for this role is give the AI permission to be honest. Tell it not to spare your feelings, and not to grade you gently just because the work is yours. That one instruction does more than anything else here.
Flattery
Too positive
Harsh
Too negative
Judge what's there
Honest & level
But honest isn't the same as harsh
Quick distinction, and it matters. Killing the flattery doesn't mean swinging the other way and getting the AI to trash everything. Telling it your work is bad is just as much a thumb on the scale as telling it the work is good.
So don't do either. Don't lead it to assume the thing is great, don't lead it to assume it's broken. Let what you put in front of it speak for itself, and have it judge what's actually there.
There's a real trap on this side too. Set an impossible bar and nothing ever clears it. You'll take genuinely good work and talk yourself into believing it's a mess. That isn't a high standard. It's a different kind of useless.
The goal was never perfection. It's holding yourself to a bar you respect. Work that's well built, well thought through, the kind of thing you're glad to put your name on. Some of what you make is already there. A good critic tells you that too.
Give it a bar to measure against
"Is this good?" is a useless question. Good for what? Judged by whom?
A critic needs a standard, so hand it one:
- Judging a marketing pitch? "Read this like a skeptical customer who skips most ads in two seconds."
- Checking an argument? "Find the weakest claim, the one a sharp skeptic attacks first."
- Cleaning up writing? "Flag every sentence that's unclear or trying too hard."
The clearer the bar, the sharper the critique. A vague standard gets you vague feedback.
Aim it at what actually matters
A critic can check a hundred things at once and do all of them shallowly. If you don't point it, it sprays.
So point it:
- Only the tone? Say so.
- Just grammar and spelling? Say so.
- Where the logic breaks down? Say so.
- Whether it's good enough to ship at all? Say so.
"Give me feedback" gets you a thin pass over everything. "Tell me the three weakest spots and why" gets you something you can fix tonight.
Feedback you won't use is worthless
A critique you don't act on is just someone's opinion. So don't ask for opinions. Ask for feedback built to be used.
That means every note comes as a full package, not a one-liner. A good critic doesn't say "your intro is weak" and walk off. It hands you four things:
"You need to do X" is a command you have to trust blindly. "Here's the problem, the fix, the reason, and your options" is something you can judge for yourself and actually run with.
So when you set up the critic, ask for all four: the problem, the fix, the reason, and the options. Skip the why and you're just following orders. Skip the options and you get one patch instead of a path forward.
See it in action
"What do you think of my essay introduction? [pastes intro]"
"This is a strong, engaging opening! A few small notesโฆ"
Translation: nothing you can use.
"You're a tough writing teacher who's read a thousand of these. Be honest, don't soften it. For the two biggest weaknesses in this introduction, tell me what's wrong, how to fix it, why it matters, and a stronger direction I could try. [pastes intro]"
Now it tells you the hook is a clichรฉ and the thesis is buried in the second paragraph, exactly what to do about both, and a couple of sharper openings to test.
That's worth more than ten "thumbs ups."
The Critic is also how you practice
Here's the part people miss.
Once the AI will tell you the truth, you've got a sparring partner. Write it, get the notes, rewrite, run it again. Draft the speech, get torn apart gently, deliver it better. The Critic turns one attempt into ten reps.
That's how the thing actually gets good. Not from one round of praise, but from a few honest passes you act on.
Takeaway
Reach for the Critic when you've made something and you want the truth about it.
Kill the flattery, but don't fish for harshness either. Tell it to judge the work honestly and let what you show it speak for itself. Then give it a bar, aim it at what matters, and demand the full package: what's wrong, the fix, the reason, and a way forward.
The bar isn't perfection. It's work you'd be glad to put your name on. Sometimes you're already there, and a good critic tells you that too.
Last of the four is the opposite of this one. Instead of judging something you made, you want the AI to make it. That's the Creator.
Want more content like this?
Get instant access to everything we offer. Upgrade to a Monthly or Annual plan to unlock our entire library of advanced masterclasses, automation workflows, and custom AI blueprints.