Skip to content
Narrative OS — The Belief Engine
🧠 Narrative OS — The Belief Engine
📐 Templates & Tools

🛠 Tool #15: Voice QA for AI

Audit, correct, and enforce your voice across AI outputs.

⁠

🧠 Purpose

To assess whether AI-generated writing (from tools like ChatGPT, Claude, Jasper, Copy.ai, etc.) is:
Aligned with your Narrative OS
In the correct emotional gear
Speaking in your defined voice system
Reinforcing the right beliefs and memory cues
This tool helps you:
Evaluate creative before publishing
Train internal teams to review AI drafts
Give feedback that sharpens future prompts
Build accountability when using outsourced or automated content
⁠

📅 When to Use

Reviewing AI-generated social posts, ads, emails, scripts, etc.
After running a prompt using Tool #10 (AI Voice Prompt Builder)
Testing ghostwritten thought leadership from AI
Auditing outsourced AI copy before publishing
Training team members to spot weak or misaligned outputs
⁠

🧱 QA Table Structure

⁠

🧪 SECTION 1: The Test Output

Paste or link the AI-generated copy here. Then evaluate it line by line.
⁠

📋 SECTION 2: Narrative Voice QA Criteria

Narrative Voice QA
Does this reflect our Core Tension?
Architecture
✅ / ❌
Notes
Is the Belief Statement echoed or implied?
✅ / ❌
​
Does the tone and rhythm match our Voice System?
✅ / ❌
​
Are forbidden phrases avoided?
✅ / ❌
​
Is this reinforcing a message from our content loop or arc?
✅ / ❌
​
Does the emotional arc trigger the correct response?
✅ / ❌
​
Is there a memory marker (line, image, moment) worth repeating?
✅ / ❌
​
Could this be confused for another brand’s voice?
✅ / ❌
​
Would our audience feel this in their body?
✅ / ❌
​
There are no rows in this table
⁠
⁠

🎯 SECTION 3: Voice Integrity Score

Rate the overall alignment from 1–5 in each gear:
Voice Integrity Score
Gear
Score (1–5)
Summary
Architecture
​
​
Exposure
​
​
Emotion
​
​
There are no rows in this table
⁠
Final verdict:
🟢 Green (Publish-Ready)
🟡 Yellow (Edit & Improve)
🔴 Red (Misaligned — Rewrite)
⁠

🛠 SECTION 4: Feedback to Improve Prompt

Feedback to Improve Prompt
What worked
What missed
Suggested next prompt
Line cadence was great
Emotion felt flat
Add silence + a memory beat
Good tension setup
CTA felt generic
Replace with “Begin here.”
​
​
“Write in the voice of The Mirror, using the phrase ‘The silence is the strategy.’”
There are no rows in this table
⁠
⁠

🧘 Closing Note

“AI won’t kill brand voice. ​Inattention will.”
This tool isn’t just QA. It’s voice enforcement at scale — powered by belief, rhythm, and emotional precision.
Want to print your doc?
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.