Does AI Coaching Work? Why the Better Question Is: Where Does It Work Best?

A recent Forbes article by Tomas Chamorro-Premuzic asks a question that sits right at the heart of the future of professional development:

Does AI coaching work?

You can read the article here: https://www.forbes.com/sites/tomaspremuzic/2026/04/27/does-ai-coaching-work/

It is an important question, but perhaps not quite the right one.

The better question is not whether AI coaching works in some general, universal sense. The better question is: where does AI coaching work best, what should it be trusted to do, and where does human coaching remain essential?

That distinction matters.

The Forbes piece explores the growing evidence around AI coaching and the areas in which it can already provide meaningful support. AI can ask useful questions. It can offer structured reflection. It can help people think through goals, habits and next steps. It can be available instantly, privately and repeatedly. In a world where high-quality coaching is often limited by time, cost and access, that matters.

But the article also points towards an important boundary. AI does not replace the relational depth, emotional intelligence, professional judgement and contextual understanding of a skilled human coach. It can support coaching. It can extend coaching. It can make certain forms of coaching more accessible. But it should not be confused with the full human relationship.

This is exactly where the most interesting opportunity lies.

At Starlight, we are not trying to replace human instructional coaches. We are trying to solve a different problem: how do we make high-quality feedback more specific, timely, actionable and regular for every teacher?

In most schools, the problem is not that leaders do not value coaching. It is that coaching is hard to scale.

Human observation is powerful, but it is also limited. It takes time to arrange, time to carry out and time to follow up. Feedback can arrive days or weeks after the lesson. It can be episodic rather than regular. Even in strong schools, the reality is that many teachers simply do not receive enough feedback, often enough, to support continuous professional growth.

That is where transcript-based AI coaching can add real value.

Starlight works by turning lesson audio into a transcript and then using that transcript to generate private, personalised coaching feedback for the teacher. The feedback is grounded in what was actually said in the lesson. It can identify patterns in questioning, explanation, wait time, student participation, misconceptions, feedback, clarity and teacher talk. It can give the teacher something concrete to reflect on within minutes.

This is not the same as a full human observation. Nor should it pretend to be.

A human coach can notice body language, relationships, atmosphere, routines, behaviour, confidence, subtle moments of confusion and the wider professional context. AI cannot fully understand all of that. But AI can do something extremely useful: it can make the spoken life of the classroom visible.

It can hold up a mirror.

That mirror can be powerful. A teacher might discover that they asked plenty of questions, but most of them were closed recall questions. They might notice that students had fewer opportunities to explain their thinking than they realised. They might see that their explanations were clear in one part of the lesson, but became rushed later on. They might spot a misconception that emerged in the transcript but was not fully addressed at the time.

These are not judgements. They are insights.

And insight is the starting point of professional growth.

This is why the future of AI coaching in schools should not be framed as human versus machine. That is the wrong binary. The more useful model is human plus AI.

A strong coaching culture might look like this:

A human coach helps a teacher identify a focus. The teacher uses Starlight between coaching conversations to receive regular, private feedback on real lessons. The teacher reflects on the reports and tries small, manageable changes. Then, when the teacher and coach meet again, they have richer evidence to discuss.

The human coach brings wisdom, judgement and relationship.

The AI brings frequency, speed and evidence.

Together, they make coaching more sustainable.

This matters because professional development too often relies on occasional moments: the learning walk, the formal observation, the annual review, the one-off training day. But improvement is rarely built in occasional moments. It is built through regular reflection, small adjustments and repeated practice.

That is the promise of Starlight.

Not to replace the coach.

Not to score the teacher.

Not to turn classrooms into surveillance spaces.

But to give teachers regular access to useful, private, evidence-informed feedback that helps them think more deeply about their practice.

The Forbes article is helpful because it moves the conversation beyond hype. AI coaching is not magic. It is not a substitute for human expertise. It is not equally suited to every coaching problem.

But used carefully, it can widen access to feedback. It can support reflection. It can help people notice patterns they would otherwise miss. It can make coaching more regular, more scalable and more grounded in evidence.

In education, that is not a small thing.

Every teacher deserves high-quality feedback. Not once a year. Not only when a coach is available. Not only when there is time in the calendar. But regularly, privately and in a way that supports professional growth.

That is what we are building with Starlight.

The future of coaching is not AI instead of humans.

It is human coaching, strengthened by AI insight.

If you'd like to see what a Starlight report looks like for a lesson in your own school, you can book a demo at https://starlightmentor.com/demo-request.

Spark Insight with Starlight today and enhance your human coaching programme with regular, private and evidence-informed insight.

🎥 Subscribe to our channel here: https://www.youtube.com/@Star21-ai

🌐 Read more on our blog: www.coaching.software

💡 Explore the platform: www.starlightmentor.com

🐦 Follow us on X: @star21starlight

The Insight Engine is written by Adam Sturdee, co-founder of Starlight, the UK’s first AI-powered coaching platform, and a senior leader with responsibility for teaching, learning and coaching. This blog is part of a wider mission to support educators through meaningful reflection, not performance metrics. It documents the journey of building Starlight from the ground up, and explores how AI, when shaped with care, can reduce workload, surface insight, and help teachers think more deeply about their practice. Rooted in the belief that growth should be private, professional, and purposeful, The Insight Engine offers ideas and stories that put insight—not judgment—at the centre of development.

🔗 Connect with me on LinkedIn: https://www.linkedin.com/in/adam-sturdee-b0695b35a/