Why are vague instructions considered the most common prompting anti-pattern, and how would you rewrite one?

beginner

Re-explain

Vague instructions leave the model to guess. Words like summarize, improve, or make it professional have no shared definition, so the model picks its own defaults.

Those defaults shift between runs and between models. You get output you did not want, and no clear signal about what to fix.

The fix is to write down what you kept in your head: the audience and format, the limits, and what counts as done.

Before:
Summarize this support ticket.

After:
Summarize this support ticket for the on-call engineer.
Write exactly 3 bullets: the reported symptom, what the
customer already tried, and the next action to take.
Use only facts stated in the ticket. If the next action
is not clear from the ticket, write "unclear".

Test it this way: could two people follow your prompt and produce matching answers? If not, it is still vague.

Rewriting in plainer words…

This answer doesn't lend itself to a diagram - it reads best . No credits were charged.

Why there's no diagram: “”

The interactive diagram is below the answer - jump to diagram ↓ · Below it, the related concept . Jump to it ↓

The diagram below the answer is the concept . Jump to it ↓

Tailored explanation · switch back to · ·

What should the new diagram focus on?

Guided practice for Prompt Engineering One question at a time. Answer out loud, get graded, see what you missed.

How well did you know this?

AI:

Saved in this browser - sign in to keep your review list.

How should your speech become text?

Listening… your words appear above as you speak - tap Stop when you're done.

Recording · cr - tap Stop & transcribe when you're done.

Transcribing with AI…

Voice:

Keep going - a few more words and AI can grade it.

Interview lens How interviewers actually use this question

Likely follow-ups

Can a prompt be too specific? What goes wrong when you over-constrain it? A strong answer notes that rules can conflict or break on odd inputs, and says each rule should trace to a real requirement or an observed failure.
Why give the model an explicit fallback, like writing 'unclear'? Without an allowed 'unclear', the model fills the slot with a plausible guess. The fallback makes missing info visible and easy to check downstream.
How would you show your rewritten prompt is actually better, not just longer? Name a fixed set of real inputs, a written pass/fail check per criterion, and a before/after comparison, not eyeballing one or two outputs.
When would you add an example instead of writing more instructions? Examples help when the format or tone is hard to describe in words. The risk is the model copying surface details, so pick varied examples.

What you can say

A vague instruction makes the model guess, because words like summarize or improve have no shared definition, so it falls back on its own defaults.
Those defaults shift between runs and between models, so you get output you didn't want and no clear signal about what to fix.
The fix is to write down what you had in your head: who it's for and what format, the limits, and what counts as done.
For example, 'summarize this support ticket' becomes three bullets for the on-call engineer: the symptom, what the customer tried, and the next action, using only facts from the ticket.
And if the next action isn't clear from the ticket, the prompt tells the model to write 'unclear'.
My check is whether two people could follow the prompt and write matching answers. If not, it's still vague, and a longer prompt isn't automatically a clearer one.

Weak answers to avoid

Says 'be more specific' and stops That's the advice, not the fix. Show a before and after, and name what you added: the audience, the format, the limits, and what counts as done.
Rewrite only makes the prompt longer Length isn't clarity. Adding 'be thorough and professional' is still vague. Every added line should remove a guess the model would otherwise make.
Blames model randomness, not the prompt The model filled gaps with its own defaults, and those shift across runs and models. Point at the gap in the prompt, then test the fix on real inputs.

Interview lens

Likely follow-ups, what you can say, and the weak answers to avoid.

Want a quick review of the fundamentals? See the Prompt Engineering cheatsheet.

← Back to all Prompt Engineering questions

Pro · $10/mo

48 of 55 Prompt Engineering answers are in Pro.

Full answers, code samples, and AI explanations that go simpler or deeper. Cancel anytime.

Full answers + code
AI explanations, simpler or deeper
1,000 AI credits / month
Cancel anytime

Get Pro → See plans & pricing →