Define 5 contexts where you'll NEVER use AI. Write them. Stick to them.

Prepare 10 deep questions before any coffee chat — converts pleasantries to relationship.

Create sheet of your 100 key relationships, last contact, next planned.

After hearing your partner, reflect back three sentences: content, feeling, and underlying value.

Apologize with: name the act, the impact, take ownership, repair step. No "sorry if...".

Each day, tell one person about a specific act they did or a precise quality — no "you are great".

Trade six escalating-depth questions with a close person to deepen intimacy in 30 minutes.

When gossip starts, use: "I prefer not to talk about absent people" — close it without lecturing.

When you err, confess before discovery — confession timing dictates the penalty.

90 days into contract: review with vendor — don't wait for renewal to find issues.

When you realize you won't meet a past promise, announce it early in a specific format — don't wait until the deadline for the stakeholder to discover it themselves.

Compare your current decision to past decisions in similar situations — are you consistent or contradicting?

Identify 5 core values and verify your decisions actually reflect them.

When sensing conflict of interest, declare it before discussion begins.

Each quarter, audit: does what I say match what I do?

Teach your child to repair what he broke instead of paying for it — integrity training.

When an LLM helps you with content you publish or send, disclose it clearly.

"AI text didn't copy anyone" is myth. Ensure you don't publish someone's idea/sentence unattributed.

Don't ask AI "write in author X's style". That = using their commercial style sans permission.