Use AI to catch hidden judgments in your observation writing
When using AI to practice observation skills, provide it with your written accounts of charged situations and explicitly request separation of observational statements from evaluative statements, using the AI's output as immediate feedback on which judgments you embedded without noticing.
Why This Is a Rule
When you write about a charged situation — a conflict, a frustrating meeting, a disappointing result — your evaluations disguise themselves as observations. "He was being defensive" feels like an observation but is an evaluation. The observation is: "He crossed his arms, looked away, and said 'I already explained that.'" The difference matters enormously: observations can be verified, evaluations cannot. Decisions built on unexamined evaluations are decisions built on your projections.
Most people cannot reliably separate observation from evaluation in their own writing, especially about emotionally charged events. The evaluation feels true, which makes it feel factual. AI is surprisingly good at this separation because it isn't emotionally invested — it can flag evaluative language that the writer's emotional state renders invisible.
When This Fires
- Writing about a conflict or disagreement for later reflection
- Journaling about a frustrating interaction at work
- Preparing feedback for someone where accuracy matters
- Any time you're describing a charged situation and want to check your objectivity
Common Failure Mode
Using AI to rewrite your account in "neutral" language. This misses the point. The value isn't in getting a neutral version — it's in seeing specifically which of your "observations" were actually evaluations. That gap is the feedback signal. If AI rewrites the whole thing, you skip the learning moment.
The Protocol
Write your account of a charged situation in full, unfiltered. Paste it to AI with: "Go through this paragraph by paragraph. For each statement, classify it as OBSERVATION (what a camera would record) or EVALUATION (my interpretation, judgment, or inference). For each evaluation, show what the underlying observation might be." Review the results. The statements you were most surprised to see flagged as evaluations are the ones where your perception is most distorted.