AI Dictionary › Prompting

Negative Prompting

Istruzioni negative (Negative Prompting)

Negative instructions tell the model what NOT to do: "no technical jargon", "avoid bullet lists", "don't mention prices", "no preamble, go straight to the answer". They complement positive constraints and fence the output's perimeter.

Definition

Using them well takes some care: models follow positive instructions more reliably than negative ones. "Write conversationally" works more reliably than "don't be formal"; "reply with only the JSON" beats "don't add explanations". Convert a prohibition into its affirmative form whenever you can.

Negative instructions stay valuable for edge cases the positive instruction doesn't cover: excluding sensitive topics, banning data invention ("if you don't know the value, write N/A instead of estimating it"), blocking recurring unwanted model behaviors.

Related terms

• Constraints (in prompting)
• Output Format
• Prompting
• AI Hallucination
• Prompt

More in Prompting

• Prompt
• Prompting
• Role Assignment
• Zero-shot
• Few-shot
• Chain of Thought
• Prompt Engineering
• Output Format

Put it into practice

• Real scenario: apply this concept with an instant AI verdict.
• Daily gym: short drills to train dictionary concepts.
• Playoff game: challenge other players, elimination-style.

Say what you mean. Get what you need.

Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.