For Executive Coacheses and Leadership Developers

Protecting Your Coaching Judgement When AI Tools Offer Quick Answers

You already know that a coaching conversation depends on noticing what a leader does not say, recognising patterns in how they describe their challenges, and sitting with silence long enough for real insight to emerge. When BetterUp AI or ChatGPT offers you a personality summary before you meet someone, or when an assessment gets scored by algorithm instead of your reading of the room, the pressure is real: use the tool or look like you are behind. The risk is not that AI is wrong. The risk is that your own judgement atrophies because you reach for the AI answer before you develop your own observation.

These are suggestions. Your situation will differ. Use what is useful.

Download printable PDF

Stop preparing coaching conversations with AI personality profiles

A Humu or CoachHub AI personality snapshot feels useful before a first session. It is not. You walk into the room already believing you know something about how this leader thinks, what they value, and where they struggle. This belief, however well intentioned, becomes a filter. You hear what the profile predicted instead of what the person actually tells you. Your curiosity becomes confirmation seeking. The executive's lived experience gets matched against a data summary instead of being met with genuine attention.

›Read only the role context and stated coaching goal before a session. Nothing else.
›Write three genuine questions about the person that you want answered, then set them aside and ask what emerges in conversation.
›Notice when you are about to reference a profile insight, and ask the leader directly instead.

Recognise what algorithm-scored assessments cannot see

Leadership assessments that use AI scoring promise consistency and speed. What they deliver is a number that misses every interpersonal dynamic that actually changes behaviour. The algorithm measures word choice, response patterns, and internal consistency. It cannot measure whether a leader's directness reads as confidence or aggression to their team. It cannot capture the gap between what someone says they value and what they actually protect in a tense moment. Your job after an assessment is not to explain the results. Your job is to notice where the leader's lived experience contradicts the score, and follow that contradiction into real learning.

›Ask leaders how their team would describe them differently than the assessment does. Trust the gap.
›Use assessments as conversation starters, not conclusions. Ask why a particular score landed where it did.
›Spend more coaching time on results that surprised the leader than on results that confirmed what they already believed.

Protect the reflective space from AI shortcuts

When a leader processes every challenge through ChatGPT or Claude before a coaching session, they arrive with a problem already intellectually solved. Your role shifts from creating space for discovery to validating or refining the AI output. This is not coaching. This is sense checking. The reflective capacity that leaders need most, especially under pressure, is the capacity to sit with confusion long enough to develop their own thinking. When they outsource that work to an AI, they train themselves to reach for external answers instead of internal knowing. Over time, they lose access to their own judgement when they need it most.

›In early sessions, invite leaders to bring the challenge as they experience it, not as they have written it down or processed it elsewhere.
›When a leader references an AI summary of their situation, ask them to describe the situation in their own words first.
›Notice and name when the room has access to the leader's thinking versus the AI's thinking. Name the difference.

Use AI to prepare yourself, not to replace your observation

Claude and ChatGPT work best when they serve your coaching, not when they stand in for it. Before a difficult conversation with a senior leader, you might ask Claude to help you think through possible reactions to a particular intervention, or to generate language for a reframe you are sensing. This is different from asking ChatGPT to draft what you might say, then delivering that language in a session. One sharpens your own thinking. The other creates distance between your judgement and the conversation. The best use of AI in coaching preparation is to stress test your own hypotheses, not to replace the work of developing them.

›Ask Claude to play devil's advocate to an observation you have made about a leader. Then decide what you believe.
›Use ChatGPT to brainstorm how a leader might interpret a particular piece of feedback. Notice which interpretations matter most to you and why.
›Prepare without sharing that you used AI. Your reading of the conversation is what the leader needs to hear.

Build a routine that strengthens your own coaching intuition

Your intuition as a coach is built from thousands of hours of noticing patterns in how leaders speak, what they choose to focus on when they are stressed, and how their stated values show up in their choices. This intuition atrophies quickly when you stop using it. Every time you reach for a tool to analyse a leader instead of trusting your read of the room, you are training yourself away from the skill that makes you valuable. A routine that protects this means regularly practicing the skills that AI cannot perform: sitting in ambiguity, noticing what you do not yet understand, and developing your own interpretations before looking for external validation.

›Keep a reflection journal after each coaching session. Write what you noticed before you check any assessment results or AI insights.
›Once a month, review three sessions where you made an important observation without using a tool. Write down what you saw and why it mattered.
›In peer coaching or supervision, practice articulating your judgement about a leader's situation before discussing what any assessment says.

Key principles

1.Genuine curiosity about a specific leader is more valuable than a personalised AI profile because it produces real listening instead of pattern matching.
2.The judgement you develop from noticing what a leader does not say atrophies the moment you start reaching for algorithmic answers before developing your own observation.
3.Coaching creates value precisely in the space where a leader sits with their own thinking long enough to change it, and AI shortcuts collapse that space.
4.Your role is to notice what an assessment cannot measure, not to explain what it did measure.
5.AI works best in your preparation when it tests your own thinking, and worst when it replaces the work of developing your thinking in the first place.

Key reminders

Before a session, write down three observations about how a leader typically shows up. Read these observations during the session to check whether they are true today. This keeps your intuition sharp.
When a leader references an AI analysis of their behaviour, ask them to describe what they actually notice about themselves before discussing the AI version.
Spend one hour a month reviewing old session notes to spot patterns you noticed without any tool. This trains your eye to recognise what matters.
Ask your most challenging coaching relationships directly: what do you think I am missing about you. Trust their answer more than any assessment score.
When you feel pressure to use a coaching AI tool because other coaches do, name the pressure. Decide whether it serves your client or your sense of being current.