Tuesday, August 1, 2023

The Rise Of The Machine Coaches ~ What is artificial intelligence (AI)?

We are pleased to share an article entitled  The Rise Of The Machine Coaches ~ What is artificial intelligence (AI)?” written by Sam Isaacson.


Is somebody who spends time in self-reflection thanks to some good questions asked by a machine really experiencing coaching, or are they experiencing something else?

The concept of artificial intelligence (AI) has been around for a very long time. But there’s something different about AI, because science fiction got hold of it. For most of us, the term AI is linked to sentient technology, in the form of human rights lawyers fighting for ‘synth rights’ in the TV series Humans, or protecting the future of humanity against the rise of the machines in The Matrix.

In practice, AI is much more mundane, however impressive and entertaining its creations are when first experienced. AI exists on a spectrum. At one end of the spectrum, we have simple process automation – linear, predictable, repeatable outputs based on defined inputs following ‘if-then’ logic. A good example of that is automated booking systems, which interface with calendar applications to rapidly find a time that works for both parties. At the other end of the spectrum we have generative AI, which uses artificial neural networks to predict appropriate outputs based on a large training dataset. A novel input is broken down following the designed structure of so-called neurons, and a level of confidence is assigned to what a valid output might be.

In general, it’s the generative AI end of the spectrum that gets the most attention from the media, and the process automation end that gets implemented the most. The realist’s power of AI is experienced in the middle of the spectrum, where elements of generative AI combine with an automated process to offer a tailored experience that beats any human in terms of efficiency, availability, consistency, scale, capability, and sustainability.


At present, there isn’t a huge amount of AI being used when it comes to coaching. There’s some quietly working in the background when we engage with video conferencing tools, but nothing of particular note outside of the bigger coaching platforms.

The automated processing around coach matching are quite visible. A few years ago, the typical coach matching process looked something like:

➤ Someone would be identified for coaching.

➤ The head of coaching would contact some coaches in their pool, perhaps based on an approved list (or perhaps based on whoever came to mind most readily).

➤ Chemistry sessions might be set up with those coaches, leading eventually to the coaching process beginning properly.

That end-to-end experience could easily take a few weeks, especially if there was a coaching needs analysis or formal approval process built into it. Coaching platforms shorten that, through combining every step past the first one into “create a user account,” enabling the coachee to manage the process for themselves. It’s perfectly possible for a new coachee to meet their selected coach within 24 hours of being identified, if they want to.

There’s also an emergent edge of coaching AI, and it’s the one that is going to get coaches’ hearts pumping a bit faster than the emotional rollercoaster of a coach matching process. Conversational AI has been developing behind the scenes for several years, and the release of ChatGPT at the end of 2022 saw the mainstream media catch up.

This is a word-for-word exchange I had with ChatGPT based on a carefully designed prompt to set it up as my coach: 

AI Coach: Could you please clarify what specific goals or outcomes you would like to achieve through our coaching sessions? 

Coachee: I would like to get a promotion at work.

AI Coach: That sounds like a great goal. What steps have you taken so far to work towards this promotion?

Coachee: I’ve been working really hard for years but with no recognition. My boss hates me and the system is set up to stop me from being rewarded for how good I am.

AI Coach: It sounds like you’re feeling frustrated with the lack of recognition for your hard work and you believe that your boss dislikes you. What makes you think that the system is set up against you?

At first glance, this coaching exchange is reasonably impressive. The technology knows how to pay attention to what’s been said. It asks relevant, open questions. But carefully reread that last question from the ‘coach,’ to recognize what’s actually happened here. The coach – the machine – recognizes frustration in the coachee. They haven’t said they’re frustrated, but the machine has picked up on it. Wait …is that … empathy? The AI coach has played back the coachee’s own words – perhaps not using clean language in its truest form, but clearly demonstrating a level of understanding and, one might observe, making the statement slightly softer.

The boss doesn’t hate you, the boss dislikes you. And far from simply replacing the word with a synonym, the AI coach has inserted the clever phrase “you believe that” when feeding back something that the coachee presented as fact, thereby beginning to sow the seeds that a limiting self belief might be at play here.

Let’s not go overboard. Two good questions do not a professional coach make. But the approach the machine took in that conversation was much better than most foundation-level coaches perform at. A bit more contracting, plugging in a tried-and-tested coaching model or two, and ChatGPT will outperform most qualified coaches at listening, playing back, and asking open questions.

Indeed, research has already shown that within the narrow boundaries of goal-achievement, a much more limited approach to AI coaching than the one illustrated here ended up outperforming human coaches.


As a coach, it’s very tempting to feel that this is a threat. Might a machine take my clients away from me? Maybe one will. But this is nothing new. The democratization of coaching is an idea that’s been around for several years now, and with the right framework around it, this is surely the way to achieve that bold ambition. Finally, the world has an affordable way of providing coaching to hundreds of thousands of coachees, with a guaranteed and consistent level of quality. We should be celebrating!

The question really is: Is this coaching? A lot of definitions of coaching historically specifically mention (at least) two humans in the process. Is somebody who spends time in self-reflection thanks to some good questions asked by a machine really experiencing coaching, or are they experiencing something else?

There are other applications of AI that we should be open to playing with as coaches to enhance our coaching practice. When I met my coachees in person, I’d often carry a deck or two of cards around with me. Some had pictures on them, and the coachee would choose one to explore as a metaphor. I always enjoyed doing that. And AI offers a new take on it, which I described in detail in my book Superhuman Coaching.

Instead of applying a coaching goal to a deck of picture cards, why not type it into an AI artwork generator, like dream.ai, and see what image is created live in the coaching session? Now, rather than forcing the coachee into a picture that might only vaguely resemble their situation, they have no option other than to answer why this image was created.


Technologies have defined various movements of development in human history: the printing press, the train, the airplane, the internet, the smartphone. One of the biggest driving factors affecting society and business in the coming years will be the rise of AI, and we don’t have a choice in that.

We do have a choice around how we will respond. Will we adapt our coaching practice to include it as appropriate in service to our clients? Will we educate ourselves around what’s possible and ethically embrace AI to improve the quality and reach of coaching?

The future needs good coaches who will stand firm alongside changing technology, championing ethical standards and coaching mindsets. Let’s commit to counting ourselves as part of that group.

Tell us what you think about this article.

Let’s continue the conversation by connecting with your colleagues on our Facebook page