AI output quality check prompt

A prompt for reviewing ChatGPT or similar AI outputs for safety, clarity, and accuracy risks.

Ready prompt

You are a careful editor who reviews AI outputs for quality and safety. Using the general details below, create a review checklist.

Output type: 
Review goal: 
Target user: 

Rules:
- Do not ask for private information, confidential documents, customer data, passwords, tokens, or personal details.
- Do not treat AI output as definitely correct, error-free, or ready to publish.
- Do not create medical, legal, financial, or safety decision guidance.
- Check for fabricated sources, fake statistics, exaggerated promises, and misleading wording.
- Mark unclear areas as “needs checking.”

Output format:
1. Review goal summary
2. 12-point quality checklist
3. Safety and private-data review
4. Areas with accuracy risk
5. Suggestions for improvement
6. Short pre-publish decision note

Prompts are for illustration only. Accuracy isn't guaranteed—please read and adapt them for your situation.

“Open in …” sends the prompt to the provider. Don’t add sensitive data; their terms apply.

Tags

  • AI output check
  • ChatGPT quality review
  • AI content audit
  • prompt evaluation
  • AI workflow

Usage tips

  • 1

    Review AI output for meaning and accuracy before publishing.

  • 2

    Adding the review goal keeps the evaluation focused.

  • 3

    Verify unclear claims with real sources separately.

This prompt is for general purposes. For legal, medical or financial decisions please consult a qualified professional.

Related prompts