R0054/2026-03-31/C002/SRC01

Research	R0054 — Prompt Claims v2
Run	2026-03-31
Claim	C002
Search	S01
Result	S01-R01
Source	SRC01

VibeSparking article on positive vs negative AI prompt instructions with research-backed recommendations.

Source

Field	Value
Title	How to Effectively Tell AI What Not to Do: Positive Constraints and Structured Output for Prompts
Publisher	VibeSparking AI
Author(s)	Not specified
Date	2025-08-14
URL	https://www.vibesparking.com/en/blog/ai/prompt-engineering/2025-08-14-prompt-do-not-to-do-playbook/
Type	Practitioner guide with research citations

Summary

Dimension	Rating
Reliability	Medium
Relevance	High
Bias: Missing data	Some concerns
Bias: Measurement	N/A
Bias: Selective reporting	Low risk
Bias: Randomization	N/A -- not an RCT
Bias: Protocol deviation	N/A -- not an RCT
Bias: COI/Funding	Low risk

Rationale

Dimension	Rationale
Reliability	Practitioner guide citing research but not a primary research publication. Recommendations are consistent with academic findings.
Relevance	Directly addresses the claim — discusses why positive instructions alone are insufficient and recommends 80/20 ratio with negative constraints.
Bias flags	Missing data concern: the cited KAIST research is referenced but not directly linked. Recommendations may oversimplify complex findings.

Evidence Extracts

Evidence ID	Summary
SRC01-E01	80/20 ratio recommendation for positive to negative instructions; converting "don'ts" to positive constraints