Use AI as an emotional distortion mirror, not a ghostwriter
When using AI to draft difficult communications, compare your reactive draft against the AI-generated neutral version to measure where emotions are distorting your message, rather than sending the AI version directly.
Why This Is a Rule
Difficult communications — performance feedback, conflict resolution, bad news, boundary-setting — trigger emotional reactions that distort your message in ways you cannot see while feeling the emotion. You overstate, understate, use loaded language, or bury the actual point under defensive framing. The distortion is invisible to you because the emotion makes the distorted version feel accurate.
AI can generate a neutral version of the same message in seconds. But sending the AI version is worse than sending your reactive draft — the AI version has no emotional truth, no personal voice, and no stakes. The person receiving it can often tell it wasn't written by a human.
This rule uses AI as a comparison mirror. Write your reactive draft first. Then ask AI to rewrite the same message neutrally. The delta between the two versions shows you exactly where and how your emotions are distorting the message. The sentences that change most are the ones where emotion is doing the writing. The sentences that barely change are your actual message.
When This Fires
- Responding to criticism that triggered defensiveness
- Delivering negative feedback to someone you care about
- Setting boundaries with someone who may react poorly
- Any message you've rewritten more than twice because it "doesn't feel right"
Common Failure Mode
Sending the AI version directly because it's "more professional." You lose the emotional signal that makes communication authentic — the parts where you show you care, where frustration reveals what matters, where vulnerability builds trust. AI strips all of that out in pursuit of neutrality. The message arrives correct but cold.
The Protocol
(1) Write the message while you're still feeling the emotion — don't filter. (2) Paste it to AI with: "Rewrite this to convey the same information in a neutral, professional tone." (3) Place the two versions side by side. (4) For each significant difference, ask: "Is my version distorted by emotion, or does it carry signal that the neutral version lost?" (5) Write the final version yourself, keeping the signal and removing the distortion.