choice Magazine

AI Coaching Comes of Age: Transformative Collaboration and Ethical Considerations with guests Sam Isaacson, Dr. Darren Leech, and David Tee

Garry Schleifer

Send us a text

What if AI could revolutionize the way coaching is performed? Join us on Beyond the Page as we explore this groundbreaking concept with Sam Isaacson, Dr. Darren Leech, and David Tee. Our conversation kicks off with the intriguing insights from their article "AI Coaching Comes of Age," where Sam shares his transition from merely writing about AI coaching to developing a real-world AI-powered coaching solution. Dr. Darren Leech explains how NHS Elect has embraced this technology, offering a fresh perspective on its application within a vast, diverse organization like the NHS.

We unpack the practical challenges and dynamic possibilities of integrating AI in organizational coaching, spotlighting tech-savvy professionals like Sue Kong who spearheaded this initiative. David Tee., a staunch advocate of evidence-based coaching, discusses the need for controlled trials to fully gauge AI’s effectiveness compared to traditional coaching. As we navigate these topics, we address common concerns about AI's role in maintaining confidentiality and avoiding bias, especially in sensitive sectors like healthcare.

As the dialogue progresses, we uncover AI’s potential to complement human coaching by filling essential gaps, such as availability across time zones and adaptability to unexpected topics. From role-play scenarios in the NHS to AI reaching new heights in coaching competencies, this episode paints a vivid picture of a future where AI and human coaches collaborate seamlessly. We stress the importance of continuous dialogue and data collection to harness AI's full potential, promising a transformative shift in coaching practices across various fields.

Watch the full interview by clicking here.
Find the full article here.

Learn more about Sam Isaacson here.
Learn more about Darren Leech here.
Learn more about David Tee here.

Gift to our listeners: Anyone can sign up for a free account with extremely limited features at https://app.aicoach.chat

Grab your free issue of choice Magazine here - https://choice-online.com

Garry Schleiffer:

Welcome to Beyond the Page, the official podcast of choice, the magazine of professional coaching, where we bring you amazing insights and in-depth features that you just won't find anywhere else. I'm your host, Garry Schleifer, and I'm so excited, I can't even say it, I'm excited to expand your learning as we dive into the latest articles, have a chat with the brilliant authors and contributors, like today, behind them and uncover the learnings that are transforming the coaching world. Take some time to join our vibrant community of coaching professionals as we explore groundbreaking ideas, share expert tips and techniques and make a real difference in our clients' lives, which is what we really want to do. This is your go-to resource for all things coaching, so let's dive in. In today's episode, I'm speaking with Coach Sam Isaacson and Darren Leech and David Tee. They are behind the scenes and the writer of the article in our latest issue What's Hot and What's Not ~ What Is Influencing Coaching, and the article is entitled AI Coaching Comes of Age ~ The Surprising Results of a Transformative Pilot Study.

Garry Schleiffer:

A little bit of each of our guests today. Sam is a coach, coach supervisor, coaching consultant, and he has a background in professional services and, of course, a digital coaching provider, one would expect nothing less. His enthusiasm is for increasing impact and visibility of coaching within organizations, with a drive to support the planet-wide coaching profession around the growing influence of technology. And just to prove that, he's the chair of the Coaching Professional Apprenticeship Trailblazer Group and contributes to several industry bodies as a subject matter expert in the field of coaching with technology, and you can see the sign behind him, which he'll talk about. He wrote Superhuman Coaching and co-edited the Coaching Buyers Handbook and so much more, and has written for choice before and been on our podcast. So welcome back.

Garry Schleiffer:

Dr. Darren Leech is a Director and Head of the Coaching Faculty at NHS Elect. In essence, a large part of his job relates to organizing, supporting and delivering coaching for people working in healthcare throughout the UK. David Tee is a coaching practitioner, consultant, lecturer and researcher and chartered coaching psychologist. He's editor of the British Psychological Society's research journal, the Coaching Psychologist, and is passionate about the opportunities AI brings to coaching. Welcome all of you. It's so great to have you here and, like I was saying, in the green room, my first time with three guests at once, so don't everybody talk at once. Okay, I was rereading this and absolutely blown away and have tons of questions about all this. But how did it get to be that it was NHS that became the organization that you chose, or got chosen, to do this pilot?

Sam Isaacson:

Well, I think that was serendipity more than anything else really, in that I I wrote a book How To Thrive As A Coach In A Digital World several years ago now, and this was before ChatGPT had become a thing, you know, before anyone was really talking about AI. But in that book I talked about this thing that in my little nerdy corner of the Internet everyone was talking about, which was called GPT-3. And I said in the book you know, GPT-3 is able to produce content that looks surprisingly human and at some point that's going to hit the mainstream, and at that point we should be expecting and actually wanting, for lots of reasons, which I'm sure we'll go into, that to provide coaching powered by AI. And so I started actively thinking, if that is going to be the case, I think coaches should be involved in the development of that and not just watching technologists come and shape the future of the coaching profession. I think coaches should be involved.

Sam Isaacson:

So, at a personal level, I was really thinking I'd like to do that myself, and I got as far as trying to create a chatbot on my own and discovered that my skills do not lie in that arena, and I came across a product developer whose skills certainly do, and his philosophy, approach to life and purpose, is extremely closely aligned with mine. And so the two of us started to develop this product, and that timed very nicely with NHS Elect, which I'm sure Darren will talk to us about a little bit more, an organization within the NHS, who were providing a conference for their internal coaching faculty, and they were really interested in, well, what's at the cutting edge of technology? Sam, would you come, having written this book and that sort of thing, would you come and share with our coaches about that? And of course, me, like an excited child with a new toy, said well, actually, that you might be interested in seeing a little bit more in depth, and the rest is history.

Sam Isaacson:

So there we go.

Garry Schleiffer:

Well, there's the lead in Darren, so tell us a little bit about NHS Elect and why you decided to say yes to this pilot in the first place.

Darren Leach:

Sure, yeah. Well, as Sam said actually, I think it was more an invited collaboration actually, rather than Sam sort of pitching to us. So we have, at NHS Elect, cut a long story short. We're a very small organization in the National Health Service. The National Health Service in the UK employs well over well a number of million people. I gather it's something like the third or fourth biggest organization on the planet. If you aggregate all of the independent and individual organizations up into one big mass. They're all quite different. They're all quite different. NHS Select is a network that organizations can choose to join if they like improvement, they want to develop their staff. One of the things that we do for those people is offer coaching. Now, our traditional market has been senior clinicians and senior folks in those organizations.

Darren Leach:

We might come back to that as a demographic we're targeting and why is it limited to them, and we currently directly employ 10 coaches and have developed, I think, a pretty decent coaching offer and service to folks in and around the NHS over the last nearly 15 years. So as Sam said, we regularly, for the 10 coaches, get together as a group and we sort out some continuing development, we sort out some reflection, we sort out some supervision and, as ever in a group of 10, it would be fair to say people have got different interests and levels of, and particularly in relation to technology, I think our group of 10 reflects the whole adoption curve, from the very mustard keen to the well let's see how this works out gang. Now, Sam was first approached by my colleague, Sue Kong, who I would put probably right at the front. You know she's in the queue for any new gadget that comes out.

Garry Schleiffer:

So is Sam so they're a match.

Darren Leach:

So she approached Sam and said look, I think some of my colleagues need to get a bit with it. Can you come and chat to us about some of these technologies and the implications? And frankly, we were a bit blown away and I put myself at the other end of that adoption scale. Pretty cynical, if I'm honest. You know, I was one of the last people kicking and screaming.

Garry Schleiffer:

Late adopters

Darren Leach:

Yeah, absolutely. So I think, probably Sue and I make quite a good combination of people to suggest to Sam. Because he was saying oh, you know, we've got this product, we're testing it in a few areas. And instantly I said, well, I wonder if we could work out how to test this, because it's not like we don't have demand. We have plenty of people wanting to access what they think is coaching. And we did give Sam a bit of a hard time about well, is this just like Siri or Alexa? Is it just going to answer their question and not actually coach them? So we tested it ourselves a bit and we're a bit blown away. We then found some friendly NHS organizations and the rest is in the article that your readers will go through, Garry. So that's the sort of background as to how our collaboration came about. And David, well, he can introduce himself. David is the completely rational scientific person.

Garry Schleiffer:

He's the researcher guy.

Darren Leach:

Yeah, who makes sure.

Garry Schleiffer:

He takes your cynicism and Sam's excitement and makes it practical and useful. Yes, there you go. There's your entry into the conversation.

David Tee:

Yes, all of that is true. My background was in the private sector and in L&D and I'd done my wonderful professional training and we were always told about the importance of evaluation, the impact of any kind of L&D activity, my own lived experience and it was echoed by my sort of community of practice was that that, for all sorts of systemic reasons and constraints, never was borne out. So you know, internal customers seeing you, just how many bums on seats have you provided this month? And we were trying to do highfalutin four levels of event and they say, no, take your trainers off of that. We need them to deliver another sort of new intake, induction or all the. So any aspiration I have to understand the effects of coaching, I recognize A). how far we've still got to go, but B) actually I think we're a lot cuter at this than a heck of a lot of other L&D activities and we should actually pause and give ourselves credit for that. So you know, I will be the first to recognize where we still don't know certain things, but there's a really dedicated effort to understand. In terms of this particular project, Sam reached out to me to say would I be interested in generating some impact data and perhaps getting some research out of this that we can disseminate? And I used to work in the tech sector and I've been very interested in the use of tech in L&D for 30 years. Very, very excited about what AI can and is already starting to bring to coaching. But also there's an opportunity here because we can program it.

Garry Schleiffer:

Yeah.

David Tee:

Same AI coach, do not halfway through the coaching session say to the client can I just take off my coaching hat and give you some advice and rescue you in the way that so many human coaches might.

David Tee:

What we're working with here is what I'm increasingly calling sort of pure or actual capital C coaching, in the way that people like John Whitmore, the whole ask don't tell mantra, setting the guardrails, and so we can really, with AI coaching, actually see what is the impact that coaching can generate. There will be people that dispute whether this is coaching or not, and maybe we'll cycle around to that during this discussion. I happen to believe that it is and can evidence as to why I think that. The interesting thing and I'm sure we'll talk about the findings is we now have emerging evidence from the work we've done with Darren and Sue and NHS Elect about the effectiveness that this AI coaching produces. What we don't know is that as good as better than worse than human coaching, all the impact data for human coaching has all manner of flaws and contaminations to it, whereas this, as I say as a researcher, is much more controllable and that's really exciting in advancing our understanding of evidence-based coaching practice.

Garry Schleiffer:

Well and thank you.

Garry Schleiffer:

You've led right into my one of my questions that, when I was rereading, and that is so was this a controlled a pilot? And I say that because they're already coaches. They knew what coaching was. Was that the intention to test it or to or like yeah?

David Tee:

I understand the word controlled in the context of randomized controlled trials. So we would have gone to Darren and say Give me 200 staff names, we'll put them in a hat, randomly pull out 100. They will be exposed to the AI coach and know they're having coaching. The other 100 will, they think is coaching but really isn't, but they'll be blind to it in the way that medical trials will give placebos. I don't know if anyone's been able to find out what a placebo in coaching is.

Garry Schleiffer:

Okay, yeah, well and okay, my fault. I talked big words to somebody who knows what the big words mean and I don't, so let me reel that back.

Garry Schleiffer:

So thank you for that, because that is kind of like a what's next question. What I was referring to is that in the article it said these were already coaches and senior leaders, so they knew what coaching was. They knew what to expect. But if you go to now what David's talking about of a controlled or a bigger study, what kind of work do you have to do in order to ensure that they know what they're getting, that they know what coaching is right? And I'm just wondering if that's kind of like your next step. Was that intentional or what is the next step for you?

Sam Isaacson:

Well, I think that is one of the big lessons that we've been taking out of. We've now been working with NHS Select for about a year probably more than a year really, but about a year in terms of delivery and the interesting thing about working with coaches is that the coaches, in a general sense, come at AI coaching with extremely strong opinions. Some of them say I'm excited about this and you know I'm right at the cutting edge. I get to play a part in it. That's really great.

Sam Isaacson:

And some say this can't possibly be coaching. You know, there's all kinds of kinds of issues with it and there are actually very few people who come to it with just an open mind. They're so curious. Maybe it will work, maybe it won't. People tend to come with strong opinions, which is different from non-coaches, for whom it's just another app that they have an access to. However, the big advantage with providing this to people who are already coaches is when they start to use the tool and it comes back with these good, open questions that are intended to increase self-awareness, most coaches, at that point, do get extremely excited because they enjoy being asked those questions, and with the non-coaches, they're the ones who at that point go, but I just want you to give me the answer. Why can't you just tell me what's going on? A lot of products that are out there calling themselves AI coaching actually do that. So the original question that Darren was raising it earlier around, Well, what's the difference between AI coach. chat and Siri or, you know, chatGPT.

Garry Schleiffer:

Right, Alexa or Google or whatever?

Sam Isaacson:

Yeah, how do I approach my first performance review where it'll just give you a nice little bullet point list of I'll do X, y and Z and AI coach. chat will say well, what's important to you about delivering a good performance review? Actually, you need a bit of onboarding to understand the value of non-directive questioning in order to make the most of it, and that has been a lesson that we've learned through these initial pilots.

Garry Schleiffer:

Yeah, yeah, well, and how do you introduce it to the non-coach, the users that don't have an understanding what coaching truly is versus consulting or therapy or mentoring? But before we get into that, you led us into and you've got your big sign there, so you got to a nswer a question for us. Tell us more about AI coach. chat. We've already talked about that it's not the same as Alexa and Siri and all that, but how does it compare to, let's say, ChatGPT, or do you put that over there too? Like, just give us some highlights of what it is.

Sam Isaacson:

Yes, so it is. If you go into ChatGPT and you give it a prompt that's along the lines of well, if you just say, act as my coach, then it will start to give you advice, because you know, the word coach is ill-defined by most of society and it just picks that up. If you come up with a stronger prompt to say act as a non-directive coach, only ask open questions, take me in the direction of a goal, it will give you decent quality, open questions. It's quite good at generating that. It will coach you according to the textbook. However, it doesn't have much direction in where it's going. It'll keep asking you open questions until it sort of loses the plot a little bit, and so what we have done in this platform is we have set up a team of AI bots that are all working together, observing the conversation and sort of acting as if they are coach mentors or coaching supervisors in the giving advice to the AI coach that you're interacting with. One of them might say I think that you should go more into the detail and commit them to action. Another might say ask them to take a step back and look at the bigger picture. Another might say use a gestalt approach in the next question, and so there's quite a mature level of understanding in the generation of then what the next question is. It takes you towards setting goals and committing to actions.

Sam Isaacson:

It's very good at performance level coaching. What we're able to do in the platform is contextualize that prompt so that it understands which organization the user is working for, why they have been given the access to the platform, and then also it assigns keywords to the conversations as it's going, so that we can then generate insights at an aggregated level without breaking confidentiality. So, knowing that 23% of people are raising stressed as a word in their coaching conversations. You don't know who has raised that word, you don't know what the context is, but you do know it's an important word. That allows you to sort of take action off the back of it, which is quite tricky to do, really, without breaking confidentiality when working with a population of human coaches.

Garry Schleiffer:

Yeah.

Garry Schleiffer:

Wow, great. Thank you for that. And I'm assuming it's a confined environment, like the confidentiality, isn't? You know? It's not like chat GPT, which is going out to the universe and asking the universe and other people can see, right.

Sam Isaacson:

Yeah, so we create a separate instance for each organization. So NHS Elect has got its protected virtual server where no other data is in it.

Garry Schleiffer:

Oh, it's even smaller. Oh, wow.

Sam Isaacson:

Of course, the conversations themselves are kept confidential. So that word is an important word in the world of coaching.

Sam Isaacson:

Actually, when we introduce technology into any coaching conversation, we have to hold it a bit more lightly because, for example, I deliver a lot of my coaching through Zoom. Well, Zoom is a large third party that needs access to my data in order to provide that service. So we're already not confidential. It might be secure, it might even be private, but confidential is not a good word to be using. When we're talking about an AI coach, actually it is the coaching client and direct with the system. There's no other human looking at it at all. It's probably more confidential than it would be with a human, and so it's a little bit of a mindset shift but the closer you get to the detail, the more comfortable I've become with AI coaching as a concept, I suppose.

Garry Schleiffer:

Yeah, well, well, let's ask the user. Mr. User, how do you feel about it? And what's NHS going to do next with if you can reveal it.

Darren Leach:

Yeah, I just pick up actually on what Sam said.

Darren Leach:

The whole information governance confidentiality thing is a big ticket item in the UK, particularly in health care. Strictly speaking, we're not allowed to use Zoom because the data is not stored in the UK. That's why another large provider is the NHS's provider of choice, because they've opened particular ways of, it's beyond my technological capability to know what I'm talking about here, Garry, but as I understand it, the data is basically not shipped abroad and no one else can access it. So we did have to jump through quite a number of hoops to get the research underway here and I think the fact it's research has helped us. The collaboration with David, our ability as a small organization hosted by a teaching organization within healthcare has been particularly useful as well. So we have been able to move things along. I think the language thing is really important. So in healthcare in the UK, certainly a number of decades ago the word coach would have been associated with remedial in medical circles. So there's some connotations, there's some understanding about that and I think our first cohort, the cohort written about, we very deliberately targeted coaches just to ask ourselves some questions about do you think they'll actually like it? Are they going to come back? And you know, obviously some of them might see it as a bit of a threat. Will they actually? Will we actually see any of them saying there was any utility or gain from using it? There were all sorts of questions for us around it.

Darren Leach:

In the second cohort, we very deliberately targeted non-coaching people and we had to set out in more detail as best we could, what they might expect, because these are people who work in a highly regulated space. They're doing risky work and therefore they're adhering to protocols, procedures, they're receiving a lot of advice, their professional development is all mentoring and I learn from expertise. So that the whole coaching model for them is quite different. Now, that said, coaching in healthcare in the UK is starting to grow. People are starting to understand what it is. They're starting to engage in it. That's also partly why we engaged in this. So we found the, I mean I'll leave David to go through the numbers and you know what people said about it but at headline level, yes, they engaged with it. I found that really fascinating. Non-coaches also engaged with it, which I found fascinating. There are some interesting who takes it up, who doesn't.

Darren Leach:

But when you think about the huge volume of people who might request coaching in our public sector health provider, whether it's a taxpayer who's paying for that support for that person or whether they're having to pay it themselves, coaching is not free. I'm yet to find somewhere where coaching is completely free in terms of time and resources and all that stuff. So, from a custodian of taxpayers' cash point of view, I was also interested in, is there a way we could scale access to coaching for a much larger number of people, because coaching to some extent, I think, has been the domain of the senior people. Those in charge get to access coaching and it's seen of a bit of a fancy thing, whereas actually some of our, some of our more brilliant people aren't getting access to this and therefore, we might not be unlocking the potential of these people.

Darren Leach:

I also thought, from a general learning point of view, to give people the opportunity to experience coaching earlier in their career and their working lives would be a good thing. I wondered if it was like a, and these are inappropriate ways of describing it, but I'll do it anyway. Is it like? Is it like having a paddle in the shallow end? So, yeah, I sort of know what water is, I sort of know how it feels and I mean, I wouldn't want to call it a gateway drug, but you never know, people who experience and use an AI coach in more research might be. Are they going to be more disposed to using an actual human coach right later down the line?

Darren Leach:

It's an emergent technology, as you've rightly said, Garry. We will find out, won't we, in the years to come. But certainly our initial experiences in the cohorts we've worked with so far, and we have had some more recent conversations about doing some more work, is the uptake is positive. People are talking about it. This appears to be a huge new technology. Why would we not engage with it? We don't want to miss the boat.

Garry Schleiffer:

Right, yeah, and you're talking primarily about the, you call them employees.

Darren Leach:

Employees, yeah, employees.

Garry Schleiffer:

And what if you were the leader in offering them to your patients?

Darren Leach:

Well, that's an interesting area, isn't it? We know that coaching approaches take them with patients who have long-term conditions.

Garry Schleiffer:

Right.

Darren Leach:

So you know again, without wanting to upset or offend anyone. In the UK, we have people who perhaps have type 2 diabetes, perhaps have respiratory or cardiac complaints. T hings that they, as an individual, should probably be managing a bit more, whereas they come to our health care institutions saying, right, what are you going to do about my diabetes then?

Garry Schleiffer:

So the employees have the coaching to push back. But now the question is what can you do to preempt the patient to start taking a look at themselves and take personal responsibility?

Darren Leach:

Yeah, so we're starting to see, particularly with a number of therapeutic areas, they are starting to take more of a coaching approach with patients who've got long-term conditions. I haven't yet seen anything about the use of AI with patients, but I doubt you'll be the first person to ask that question, Garry.

Darren Leach:

But the reason I'm not mostly focused on that is because our job at NHS Elect, is to work with the employees and the staff. But we are coming across clinical employees who, you know, when we teach our fundamental coaching skills courses and things, they say, oh yeah, we're doing some stuff with this grow model, we're starting to work with this. You know, I can think in Hertfordshire and various other areas of our country people are already using that. So if they're doing it in real life, why not a bit of AI in future too? So to me it's just a further research question.

Garry Schleiffer:

Yeah, well, Mr. Researcher, I think the gauntlet was thrown over to you and you see antsy to jump in and add something.

David Tee:

So I think Darren hinted at me running through the numbers. So if I quickly do that, and I know that we touch on this in the article in choice magazine as well, but for people who are jumping into the podcast and haven't had a chance to read it yet, we wanted to expose people to sessions with this AI coach and obviously it's up to them what topic they choose and what their particular goal is, and that's going to vary from person to person. So the success criteria, the outcome measures, needed to be relatively generic and I look to what typically in coaching are considered relevant factors and to what extent do they feel they've made progress with their goal. And I know you get people saying I think we exaggerate this importance of goals in coaching, but I've had the privilege of doing a number of national and international surveys where it's around 95 percent of coaches say they do work with goals. So I think actually, boots on the ground, goals do play a very, very, very key role. So I can sleep easy with that. And the other measure is what's called generalized self-efficacy. So my ability to think. If I put my mind to this, I reckon I could learn to touch type or juggle chainsaws or whatever this thing is I can't do now, but I want to be able to do at some point in the future. So my general sense that putting effort in is going to give me desired outcomes.

David Tee:

So those are the two things we were interested in. Across the two waves of it, we've had about 80 people take part and provide us with a full set of data, predominantly female, mid 40s, which Darren tells us is very typical of the employment demographics of people that work in the National Health Service. We did also give them this suite of presenting topics, sort of going into it. What do you think you want to use the coaching for? I was hoping that they would make quite clear choices. Most of them picked about four or five out of the possible seven. And looking across the two waves of research that we've done, so this is including new data beyond that in the article, so this is an exclusive for your podcast audience. They all tended to be picked by around 55% of respondents.

David Tee:

So it's things about working with immediate challenges at work, raising my self-awareness, thinking about my career momentum, me as a leader, me as a communicator very, very familiar territory for any coaches who have built up some experience working with their client base. But if I start with the coaching itself, taking the averages, they were saying they were already at about 48% progress with whatever their goal was. They weren't coming in, so I've got this brand new thing. I've done nothing with it at all. Give it your best shot AI coach. It was often we've got this AI coach, would you like to play with it? And I think some of them were driven by curiosity rather than I was just about to reach out for a coach, let's try this. My parallels if you brought people into a room and said there's a virtual reality helmet, who wants to put it on and have a play? It was that kind of thing. So they were drawn to it, but it was often something they were already working on. After the coaching they were reporting 59% progress. So there's an uplift of 11%, and I hinted at this earlier. We think that's wonderful in and of itself.

David Tee:

We wish we knew if it was as good as with a human coach.

Garry Schleiffer:

Well, there's your placebo or whatever, there's that controlled environment.

David Tee:

Yeah, and the other thing that excites us is that this does compound, so people that came back for more sessions reported greater progress on their goal, which is what you would want. So that's cool.

Garry Schleiffer:

Yeah, but by then they're like they bought in.

Garry Schleiffer:

Darren's, you know, Mr. Last Adopter is now at the front of the line going. Yay, let's go.

David Tee:

And the other one about self-efficacy. Again, average scores before the coaching came out 65 percent using the generalized self-efficacy measure, afterwards 70. S we're seeing an uplift of five percent and I go away feeling confident from the session. I can do perhaps more with this than I was telling myself I could anyway. So that's great. I would love there to be more people, because the more participants, more confidence in the data, but then what researcher wouldn't want that? So it is early days with the technology. It's early days with AI coaching and the technology, but the initial findings and I'm supposed to be an objective detached scientist, but there's a part of me, Garry, that can't help feeling very excited as well.

Garry Schleiffer:

No, it's hard to tell, really. The man's bouncing in his chair.

David Tee:

I'm lousy at poker.

Garry Schleiffer:

You know, there's something that I'm curious about and I think you started the conversation, Sam, when you talked about a marker of, let's say, stress. So when you find these, so first of all can you see what they're actually choosing and can you then turn around and take what they're choosing and advise the next cohort or other people, this is what you get out of coaching? So you're taking the information and marketing it back to the next cohorts in the pilot.

Sam Isaacson:

Well, I think for the research projects that we're talking about specifically here, that hasn't been possible because, in order to satisfy the requirements around data privacy and security, we have not been recording any of the conversations. So, in the generics of off the shelf platform, then the conversations are recorded so that, as an individual, the next come back in. If you said I'm going to talk to my boss, one of the questions it could ask you is you said you were going to talk to your boss, one of the questions it could ask you is "You said you were going to talk to your boss. Did you actually go and do that? So it holds you to account a little bit, and that hasn't happened here, so we can't really comment specifically on that. The analysis of the keywords can inform wider strategy in all sorts of ways, and so one of those absolutely ought to be "well, you know, we've used the platform for six months and this is what people are showing the most interest in, so perhaps we should create a custom conversation that is about that topic and it's designed to address that. So what's quite interesting, I think, we did not have any direct influence over the research questions that were being asked. I just trust Dave. He knows what sort of thing we should be looking at, and he'll go off and do that.

Sam Isaacson:

What was in my heart was I can never control what a coaching client turns up and says to me, so I might say at the beginning "you're on a leadership development program, what would you like to work on today?

Sam Isaacson:

And they say, "I'm having a really hard time with my husband, can we talk about that? And my response as a coach, professionally, is yes, yeah, of course. You can't just say to an ai coach well, you're here to talk about this, though, because this is the content that you're here. So my heart with it was I want it to be able to go anywhere that you go as a user, and I just believe that good coaching principles are effective, and then let's just hope and see if the research demonstrates that. The platform does note your goals. It is designed to note that, but it isn't specifically designed in a way that is going to increase your self-efficacy. However, if you're asked open questions and the AI coach has been told to believe that you're capable of doing it, there is this sense of osmosis, that you start to pick up the belief yourself, and I'm very pleased, of course, to see that that is the result that came out.

Garry Schleiffer:

Well, good. Good to know. And so kind of piggybacking on what Darren was saying about being a late adopter. So I am, and you and others you, I think you too, mentioned as well, Sam, that people are afraid of it and it's usually what they don't know. We talked about this in our previous call. I'm not one of those people. As a result of doing that issue on AI and coaching and talking to you and everyone else, I'm like I'm looking under the hood, I want to know what's going on. So I'm in the process of creating a Coach Garry AI coach, with an organization that I met at a conference, amigo. ai, and so I'm just like I'm motoring ahead and thinking about it. This has been very helpful for me too Confidentiality, common topics, how do you maintain confidentiality and still, you know, find out what you can market to people to let them know what they can get out of having an AI coach.

Garry Schleiffer:

But anyway, that's all just about me, so we'll stop that. In working with some friends of mine who are just as biased and fearful about AI, some questions came up and you mentioned one of them in your article and t heir question, is can an AI coach cause transformation?

Sam Isaacson:

Well, I've noticed recently. I go to a lot of coaching events and conferences and webinars and things which has got lots of coaches in and aren't tend to be talking about AI, and it appears to me, this is going to sound horribly judgmental through a generous lens. It feels like when coaches gather to talk about technology, there's a bit of a competition between us as to who can say the most comforting thing to us as coaches, that it's all going to be all right. And so some of those comments are oh well, AI can never do what humans can do because there's bias in data sets, or because we're concerned about confidentiality, or because there's an impact on the environment, or because it can't read body language or whatever else it is. And this idea of, oh well, AI coaching is good at transactional coaching and human coaching is good at transformational coaching feels to me like another of those little throwaway statements which oversimplifies something which is so incredibly nuanced. We find it difficult to even define what the word coaching means, and so the route that I take is to explore the idea that, while I agree that AI coaching is coaching, we already see there's lots of different versions of coaching in the world.

Sam Isaacson:

AI coaching is not the same as human coaching. They're different things. So, for example, an AI coach cannot sit with you and be present in a moment of silence. It can sit very quiet. Now it will sit there quietly with you but it isn't there with you. It's not feeling that moment, and just the fact that it isn't means it hasn't got the same sort of power.

Sam Isaacson:

However, an AI coach is going to be there at two o'clock in the middle of the night when I wake up worried about a meeting and I just need a coaching session now. And to all my clients out there, I apologize. You can try and phone me as much as you want, you're not getting through at that time. Also, an AI coach is able to hold a conversation open over multiple days. We had a really interesting piece of feedback where somebody had opened a session on the Monday, got halfway through a conversation, left the window open in the background while they got on with the rest of their week and then were closing their windows down on the Friday, saw their half finished conversation and that was the moment that they went, Yes, that is right. This would be the right answer to that question now I've had several days of it processed unconsciously and again, that isn't something a human can do, so it's not the same.

Sam Isaacson:

In some ways, I think it's probably better, in some ways it is definitely worse. If we as humans, I think I said this last time, if we're trying to compete against an AI at being cheap, at asking open questions, at following a coaching model, we're going to lose that competition. We shouldn't try to compete. We should be recognizing that it's present and then seeing how we respond.

Garry Schleiffer:

Yeah, thank you.

David Tee:

I do strongly think that is the wrong question. Can we get AI to be as much like a human coach as possible? I think they should end up as complementary modes of coaching. I think that there are, as Sam alluded to, things that humans will always be better at, and there'll be things, and we'll probably emphasize these more in this conversation, that an AI coach will always be better at. And I would add to the many examples Sam gave about being genuinely non-judgmental and the client being able to believe that my coach isn't judging me, even if they tell me that they're not. I might think they're secretly approving. They're not going to think that with an AI coach. There are different things. The analogy I often use as a vegetarian, and I don't know what it's like where you live, Garry, if I go to the supermarket and I want to buy vegetarian food, it's all shaped as a burger or a sausage or I can't believe it's not chicken or shrimp.

David Tee:

I don't want chicken or shrimp. Why is it trying to pretend to be the very thing it isn't?

David Tee:

I think it's the same with AI coaching.

David Tee:

It's the wrong question to be asking. Let's recognize what these are good at, let's target them in a very sensible way where they are going to create the most value and the most impact, and lean into both of them in service of our staff and our client organizations.

Garry Schleiffer:

I hear another really great article by you guys. Just thinking. Help the coaches. I don't want to say embrace, but fully understand. Maybe what you said David was to complement. How to see AI as a complementary, augmented situation rather than a competitive situation. I'm actually building my offering as a collaborative complementary offering. Just AI coaching, just one-on-one and a combo one because I still feel that they should have that opportunity to speak with me as a one-on-one coach to maybe like recap or go, you know, did you get everything you wanted, kind of thing, like that sort of thing. Gentlemen, we're running short of time. Thank you, but not yet. I do want to ask you and each of you, maybe briefly this time, because I love hearing everything, but unfortunately, what's next for AI and coaching?

Sam Isaacson:

Well, I'll try and keep it short, who knows.

Garry Schleiffer:

I know, that's a huge question, but just in your front screen.

Sam Isaacson:

So, absolutely we want to improve the consistency and quality of the coaching conversations that are happening and also explore, well, if an AI coaching conversation that follows something that you might recognize as being similar to some sort of grow model, what could sit adjacent to that that would make the experience overall more effective. So, for example, could we build in a role-playing exercise so that when you say I'm struggling to think about delivering my first performance review, the AI goes all right, we're changing what the screen looks like. I am now the person you're delivering the performance review to and I will act like that, and that feels like it can do it in a bit of a more authentic way than a human, something like that. So I think, expanding out from a pure conversation to something that makes the most of the technology.

Garry Schleiffer:

Well, I'm here to tell you, I work for Platform and they have six of those scenarios set up already on six different topics and they're specifically AI role play scenarios exactly like you described. So it's out there and that would be a good thing, for you know, NHS. I don't know which side of the screen you're on over there, but you know if the information were to be shared or tested, that sort of thing. And Darren, what are your thoughts about from perspective of the lens NHS and AI coaching.

Darren Leach:

I might reverse the question, actually, Garry, and say it's not what's next for AI, because I think AI is not an "if now is it, it's just a when it is happening. Basically, we've got to accept that.

Darren Leach:

So I think it's what's next for human coaches, because we've got to learn to collaborate and we've got to learn to have AI and actually other technologies to complement and enhance what we do. And if we can make our coaching episodes with our clients five, ten percent more impactful by utilizing technologies which might include AI, why would we not want to do that if we're working in our clients?

Darren Leach:

To me it would seem crazy. I guess the other thing, from a really purist coaching point of view, which Sam and David know I have views on, is well, there's some challenge here for coaching supervisors, isn't there? Because are you going to start supervising the AI coach? How's that work?

Darren Leach:

And how do you supervise coaches who reject AI in the 21st century?

Darren Leach:

But anyway, I'll leave it there.

Garry Schleiffer:

Wow okay, well, definitely don't have time for all of that and great questions for our listeners to think about. What corner are you in on this? Where are you on the on the spectrum? Final words, David?

David Tee:

So one very quick point which I alluded to is I hope that as it penetrates more and more organizations coaching solution mix, that they'll start boxing clever with it and using it to complement rather than seeing it as a threat and actually saying what its unique abilities are. But the other thing in terms of what's next for AI coaches themselves separate to the wonderful work we've been doing with Darren and Sue in NHS Elect, I was involved in another project where we got a bunch of ICF assessors to say treat this as if it was a human coach and assess it. That was summer of 2023. It pretty much crashed and burned. It was actually then this summer 24, and it demonstrated the ACC, the Associate Credential Coach competencies, and it was already starting to pass some of the PCC Professional Credential Coach competencies. That's in 12 months. Let's think ahead to summer 2025 and maybe it's starting to show some MCC level ability. So it is cleverer and cleverer.

Garry Schleiffer:

If you don't know David Drake, you need to talk to him, because he has a whole conversation about competencies versus masteries beyond the PCC level, and I'm going for MCC, so would you get it all done with ICF before I arrive at the door? Oh my gosh, gentlemen, thank you so much. What would you like our listeners to do as a result of this conversation and the article?

Sam Isaacson:

I think my ideal, I think it's probably the idea of everybody on here really, is we'd love to keep this conversation going and expand the dialogue, and so we want to discover more about what is possible. We'd like to gather more data from a larger number of participants across a more diverse range of organizations, contexts, countries, and see, you know, what more we can learn. And, like saying right at the beginning, one of the little things that you said was something along the lines of we're here to increase the impact of coaching in the lives of our clients. What a lovely way this is to be able to reach such a large number of people in a really practical way, and so, yeah, it'd be great to get more people on that journey.

Garry Schleiffer:

Yeah, no kidding, did he say it all for you guys too?

Garry Schleiffer:

How can we reach you, Sam, gentlemen, should anybody have any questions?

Sam Isaacson:

I think I'm certainly on LinkedIn and if you go to aicoach. chat then you'll be able to see the website there. Create an account and have a little play for you.

Garry Schleiffer:

Awesome, thank you. Thanks again. This has been amazing. I love these kinds of things because I'm a rah rah. I'm declaring my bias right away. I'm a rah rah on AI, so I have to kind of tone it down and be a little bit more for my listeners because they're on both ends of the spectrum. But it's been very informative. Thank you so much for your contributions. Seriously, as a member of the coaching profession for over 24 years, thank you for everything you're doing. Thank you for the transparency, the wisdom and the foresight to keep it going. Thank you all.

Sam Isaacson:

Thanks Garry.

Darren Leach:

Thanks Garry.

David Tee:

Thanks Garry.

Garry Schleiffer:

That's it for this episode of Beyond the Page. For more episodes, subscribe via your favorite podcast app, most likely the one that got you here. If you're not a subscriber to Choice Magazine, you can sign up for a free digital issue by scanning the QR code in the top right-hand corner or by going to choice-online. com and clicking the Sign Up Now button. I'm Garry Schleifer. Enjoy the journey of mastery.