What Happened When I Talked About AI at International Coaching Week Singapore

I did not go into the room expecting consensus. The session was called “The Augmented Coach: Amplify Your Voice with AI” and it was designed to do two things that do not always sit comfortably together: unsettle and inspire.

It did both. What I did not fully anticipate was how much the room wanted to talk.

The opening that set the tone

The session opened with a provocation. Not a question, not a poll: a scene. An AI holding complete attention for twelve minutes without once needing the client to like it. Without checking whether it had said the wrong thing. Without the ambient self-monitoring that runs underneath so much of what we do as coaches.

The question I asked the room was simple: what did that do to you?

That question stayed live for most of the fifty minutes that followed.

The part that triggered people

The middle section of the talk did not let coaches off the hook. It named the risk that most AI conversations in coaching circles avoid: that the biggest threat to our professional ethics is not malice but convenience. That the boundary between appropriate and inappropriate use of AI tools erodes gradually, through habit, not through a single bad decision.

The research on this is striking. A significant majority of people who worry about sharing sensitive data with AI tools do it anyway. Coaches are not immune to this dynamic. If anything, the intimacy of what we carry — client vulnerabilities, organisational politics, developmental hypotheses about people’s shadow sides — makes the stakes higher than in most professions.

Naming this directly created discomfort in the room. That was the intention. A provocation that produces comfort has not done its job.

The part that opened something up

The second movement of the session was different. Three live demonstrations of AI used not to replace coaching judgment but to extend it: a Hypothesis Challenger that generates alternative explanations for what a coach believes is happening with a client, including at least one that implicates the coach’s own blind spots; an Interference Detector that reads a session transcript for moments where the coach stopped following and started leading; and a Longitudinal Mirror that reads across months of session notes to surface the underlying narrative a client is unconsciously constructing.

Each demo was framed the same way: this is AI as the supervision you cannot always access. Not a substitute for a human supervisor, but a thinking partner available between the sessions when supervision matters most.

The room shifted here. Curiosity replaced defensiveness. Coaches started asking about their own practice, their own patterns, what they might see if they looked that closely at their own interventions.

The ending that stayed with people

The session closed with a polarity: Human Presence and AI Augmentation. Not a problem to solve but a tension to manage — two poles that need each other. Coaches were invited to physically voice each quadrant in the room: what we gain from full human presence, what we risk losing by over-relying on it, what AI augmentation genuinely offers, and where it goes wrong.

Something shifted in the room during that exercise. The conversation stopped being about AI and started being about practice. About what we are actually doing when we coach, and what we owe the people who trust us with their development.

The questions continued for days afterward. In the corridor, by message, from coaches who had gone home and started thinking differently about tools they were already using.

That is what a good provocation does. It does not close the conversation. It sends people back to their own practice with a question they cannot ignore.