Catherine McDonald  01:51

Hello and welcome to this episode of the Lean Solutions podcast. My name is Catherine Macdonald, and today I will be talking about the AI readiness gap and how to close it. So many organizations are investing in AI, but few stop to check if they’re actually ready for it. Our guest today is Diana Acapar, and she will be helping me to explore this AI readiness gap. Welcome to the show, Diana.

Dajana Achelpohl   02:23

Thank you so much, Catherine. Really looking forward to this chat.

Catherine McDonald  02:27

Yes, me too. Me too. We know each other a little while from LinkedIn, don’t we? Yeah. So what? So I know you’re an AI expert, so you’re the right person to have on the show today. So just by way of explanation for our listeners, the gap that we’re going to be talking about this gap is the difference between basically just having access to AI and then being able to get value from AI in your organization. And so, what you’re going to do with us, Diana, is you are using five simple questions. You’re going to share a practical way to assess readiness among people, processes, leadership, and organizations. So, those of you listening, you can look forward to leaving with a simple AI readiness check that you can use immediately. Am I right in saying that? Absolutely,

Dajana Achelpohl   03:14

you summarized that very well. I couldn’t have said it better myself.

Catherine McDonald  03:18

Okay, and we will, we will, you will be doing that, but you will also, we will also be having a chat beyond that as well. So before we go any further, Diana, I’m just going to introduce our listeners to you, if that’s okay. So Diana Akapor is the founder of AI Change Maker, where she helps leaders and teams adopt AI in a practical, people-centered way. She spent nearly 20 years in big tech with Google and PayPal in leadership roles across sales, operations, and service management. And today, she works with organizations to build AI literacy, identify meaningful use cases, and make AI adoption stick, which is a super big job. And you also mentioned earlier that you lecture in Trinity as well, so you are you are a busy lady, but it’s great. It’s a it’s a very interesting space to be in at the moment. So maybe you can kick us off, Diana, and tell us a little bit about this whole readiness gap and what we need to do to actually make sure that we are making best use of all the AI we have out there now. So, can you tell us a little bit more about the gap and what we need to do to close it?

Dajana Achelpohl   04:29

Yeah, absolutely, love to. I think you’ve already set it up quite well. What we’re really talking about is that having access to AI is not the same as getting value out of AI, and I think a lot of organizations start experiencing that lately. I see a lot of organizations that have bought access to the tool. Microsoft Copilot is very commonly the tool that has been bought. The licenses that are there, and now people are waiting for value to show up. And it doesn’t. It doesn’t just magically happen. And that’s really what I’m talking about when I say this: this gap. So people do have access to the tools. They have done the spending on having access to these tools, but they’re not getting that value. So what we’re really looking at is that gap between we have an AI tool, but won’t just magically get us that value. What do we need to do to get there? I think all of the hype around AI has a lot to answer in this regard. I think was very much suggested that AI is somewhat of a silver bullet. You get the tool, and this will magically work for you. And now we’re starting to experience that it won’t. I see this in all size of companies. This phenomena, both in smaller organizations and in bigger ones. And I guess a few symptoms to look out for to see if if you are caught in this gap are a few things like there’s loads of AI activity, but very little measurable results. You see a lot of scenario where it’s mostly individuals using AI. You might see good use for individual productivity, which is amazing, but might not actually get you far as an organization. You might see very fragmented use across teams, but nothing really that is creating a cohesive picture there, so a lot of silos, a lot of things that are potentially being duplicated, and then I also see a lot of pilots. I do talk to organizations that have 25 AI pilots on the go at the same time, and some of these pilots have been running for a year plus, and they are still hoping that that somehow this will magically sort itself out. So they are the kind of symptoms, I suppose, I see where then can very clearly say you are stuck in that gap, and we need to address that.

Catherine McDonald  06:51

Yeah, and you’ve just answered my next question, which was, well, what exactly is going wrong? Where are we going wrong? But you’ve just, I think, especially the whole pilots and initiatives and everyone’s time taken up, which is great. Organizations are doing the right thing by letting people experiment right with AI. We have to, but there, you know, I think there has to be some shape, some structure, some measures. Why? Why are we not able to get those and and get some sort of measure of the work we’re doing. I find that hard to understand.

Dajana Achelpohl   07:26

I know, and I think there there is an element there, and you and I talked about it before the the recording, where again people feel that a lot of common business sense is not needed for AI, and again because this was sort of sold and suggested to us, you still need to put in a lot of the groundwork that you would put in with any other new initiative, any other change you’re bringing to your organization. And if you do not do that, this will show, right? And it would-it’s certainly something you’re talking there about measuring. I talk to so many organizations who are actually very unclear on what a good result would look like, which is a problem, right? If we don’t actually know what measure to look at to figure out if our AI rollout is successful, we have a big, big problem which we need to address.

Catherine McDonald  08:17

Yeah, yeah, and it just it reminds me about what we talk about in in lean, so I know you have some knowledge of lean as well, Diana. But some listeners on the show would be familiar with some phrases, I suppose, that we use in lean, like current state. Understand the current state before you go on and try to create a more improved or future state. And it’s like that part was just both parts were just completely skipped because of the hype that you talk about. There, there, this this measure of the current state, this this analysis of the problem. What are we trying to solve? What are we trying to make better? Why? What are we going to prioritize? It’s like it was just totally everything we need. Our organizations know about how to be lean just went out the window because of this big AI hype. So hopefully we’re learning our lessons now, and we’re just taking a little step back. Can you see? Can you see things changing in any way? I

Dajana Achelpohl   09:13

can, and I can definitely see organizations that are, as you rightly say, they are taking that step back. They’re looking at the bigger picture, and they want to do this right. So I also want to assure your listeners that this is not something you can’t get out of right. This is actually something you very much can get through this gap and to get to the stage where you are actually going to see real value from it. I think another interesting phenomena that that fits into that whole gap scenario is if you look at usage of AI tools after they’re just being launched, you usually see a lot of excitement there, a lot of experimentation, and then that literally goes off a cliff. There’s very interesting stats that show that three weeks after Microsoft Copilot. It was launched in an organization. Only 20% of people still regularly use it, which is a real concern. I’m getting like nightmares from this. All of these licenses, all of these seats that are being bought somewhere, and nobody using them. So it’s definitely something where organizations are now waking up to the fact that they need to do something about

Catherine McDonald  10:21

yes, good. I think it was maybe born out of oh well, we can’t fall behind. Everyone else is getting it. We have to get it. So it was just one of these things. Yeah, yeah. A lot of

Dajana Achelpohl   10:30

FOMO, absolutely. A lot

Catherine McDonald  10:32

of FOMO, definitely. Okay, so you have a nice method for helping organizations, I suppose, understand where they lie in their readiness for AI. You you you call it your your five questions, isn’t that right? Yeah. So can you yeah can you maybe just talk us through what they are? I know I know you try to keep it simple. I love simple. I think everybody

Dajana Achelpohl   10:56

absolutely

Catherine McDonald  10:57

needs to come away with simple from this. So what what are the five questions, and how do how do you ask them to to understand your AI readiness?

Dajana Achelpohl   11:06

So, the five questions are really simple. And initially, when I say this to people, they’re kind of going, “Dah, I could have figured that out myself. But that for me is a good result, right? We want something that’s intuitive where people immediately get it. So, the five questions I suggest organizations ask themselves is why, how, who, what if, and then what. And let’s maybe go through those one by one so that we know what we are asking in each of these questions. And I always suggest organizations go through all five of them, they are all important. It’s good to get the full picture. You might have a good answer to one of them, and then that’s good news. There might be other areas where you need a bit more work, right? So we are starting with the very simple but very hard to answer question of why. Why are we actually bringing AI into the organization, and we were talking about organizations being worried about being left behind there earlier. Your why for AI can’t be everyone is doing it. So we need to start with a really good reason. Why are you bringing AI into your organization? What is the problem or opportunity you’re looking to address. What are your overall business priorities, and what role does AI have to play in that one? Sometimes that is an incredibly hard question for organizations to answer, especially if they have already made the investment and now they’re getting asked these uncomfortable questions. There’s no shame in that. I think it’s something that you need to think through, something that you need to work through, and the answer doesn’t have to be overly complicated. We have AI to support our business objectives. Of X is a good starting point, and then we can go further from that. But the first big question is the why. Having a good answer to that question will also address one thing that’s really important: your team, your employees, your staff needs to understand the why. So you need to actually have a really good answer for that. Otherwise, it’s very unlikely that they will adopt this, or which of course is a big thing with AI. They’re going to be worried. Are you bringing AI into the place out of us? I regularly hear people telling me how they’re concerned about AI coming into the workplace. A lot of people asking these questions: Am I just training a system that will ultimately replace me? They’re very valid questions and concerns, and I think by having a good answer to that why questions, companies can actually show how AI fits into their business goals. And of course, one of the reasons why you bring in AI might be that you are looking to do things faster, potentially with less staff, and again, that is absolutely legitimate. But your teams deserve to understand that. So the why is very important, and AI, as we said, can’t be the objective in itself. Right? You you need to have a work element that you want to improve something you want to drive with AI. So that would be the first question: the why.

Catherine McDonald  14:28

Yeah, excellent. Yeah, I completely understand what you’re saying. So I know I’m working with some clients at the moment, and some we’re trying to maybe in in a look at these AI receptionists or AI just just just little bits of AI here and there to take away the manual work from some of the frontline staff, and I do know and notice some resistance there. Not not in a massive way, but I just I can feel and fence people’s apprehension to well what. Will this mean for me? Almost to the point where they they just don’t want to go near the AI. They don’t they don’t want to know about it. So, what really helps there, and you might have found similar, is just like you said, helping people to understand. Well, if we do have this AI, what are the things I can do and I can get in my job that I haven’t been able to get to to this point, so I know one of the the people I was working with, she talked about just getting time to work on the the strategic stuff. Like she’s quite she’s she’d be a senior manager, and she she’s caught up in the do do do of work. Whereas when she you know when we started to look at this AI and and and bringing in a little bit more of the AI to do more of the manual work, she could see then that actually this is the purpose-not to replace me, but to free me up to do this kind of work. So I think making that very clear is important. What are we talking about when we say AI is going to free us up to do more important work? You know, name it. I think that’s huge. Yeah, you need to be

Dajana Achelpohl   16:00

able to paint that picture needs to become very clear for people what’s in it for them. How is my life going to become better through this? And very often we do have that scenario where people are absolutely bogged down with all of the admin work, work they don’t enjoy, work they don’t want to do, but that needs to get done. They don’t have time for the things that excite them that they want to do more of, so if you can show them, hey, AI might actually take that off your plate, so that you can spend time on the things you really want to do. That’s a much better conversation and much more engaged workforce, which you will have as a result.

Catherine McDonald  16:34

Yeah, really important starting point. So that’s the why. So the next the next step is how. The

Dajana Achelpohl   16:40

next one is how, and we’ve kind of again touched on that as well. So to be able to answer the how, we need to understand the work because we want to understand here how AI would actually fit into the work. That again is something that requires, in many cases, a bit of work. So we need to really understand how work flows in an organization. We need to understand input activities, handoffs, etc. Again, something where there’s no doubt crossover to to lean. Right? We need to find those bottlenecks, repetitive tasks, all of those things. I think really important in at that stage when we try to answer the how that you might have to involve people that are close to the work. Sometimes I do this work with the C-suite in an organization who wants to bring AI in, but they might be too far removed from the actual work to be able to credibly answer that question, so that might have to be an exercise where we are bringing in people that are closer to the work, where we are running focus groups, or we are going out there to understand that. But I can’t overemphasize how important this is, right? If we don’t understand the work, we will not be able to figure out how to integrate AI meaningfully into that workflow. If AI is something that sits beside the way work actually gets done, people would stop using it very, very quickly. So it’s again something that might take a bit of time, might be a bit uncomfortable, but it’s absolutely worthwhile. We need to understand how the work actually flows, and able to then be able to identify where AI might help.

Catherine McDonald  18:32

Yeah, I agree with that as well. Actually, speaking to some people lately about their approach, their lean practitioners, lean consultants, lean leaders and organizations, and back in the day when we studied lean, we were taught about things like the eight wastes or start with understanding, you know, the problems or the inefficiencies. AI is kind of changing that as well because it’s now no longer just about oh here’s our process, let’s find the waste. It’s actually about looking at it from a more proactive opportunity point of view as well. So something might not be very wrong. It just might be taking us time, and we might not know that that’s wrong. But when we factor in where AI can help us in each of the steps of the process, we might very quickly then start to see much better ways of doing things. So even the way that we look for problems and look for opportunities, I think, is changing. However, the the the fundamental piece that you mentioned is remains. We have to work with the people do doing the work to actually collaborate and understand where those problems and opportunities lie. So, 100% agree with me.

Dajana Achelpohl   19:42

And that people element so important, right? I have worked with organizations where leadership has assured me this is how the work flows, and then it took me about 20 minutes talking to somebody who’s actually close to the work to then figure out that that was not at all how the work. Getting done, you know. So again, this is not new. This is not unique to AI, but AI will not work if you’re not looking into this properly. Yeah, yeah. So that’s the second one, the how. The third one then is the who. So the people are obviously incredibly important in AI. So we really need to understand who needs to be involved, who needs to be ready, who needs to have the right skills to make our AI approach work. So we might ask questions like, “Who owns this process? Who does this work at the moment? They need to be involved in this. Also, I always want to include the leaders in this, I think, leaders understanding AI, modeling AI is is very very important. So we want to get a good understanding of who is involved in this work. Often, I see that the work obviously doesn’t stop at departmental boundaries. So we might have a scenario where there’s part of an organizational function that says we want to do AI, but then it turns out that there’s another function involved, right? And again, even understanding that, getting those people potentially involved is also important. So we want to identify these people that are involved to both be able to partner with them to redesign the work with AI, but also to know who we need to upskill. AI, as we said, isn’t as self-explanatory, despite all the hype messages we’ve heard. So we are going to have to invest in upskilling these people. So they’re going to need a basic level of AI literacy. They need to understand how AI works, why it’s good in certain things, not so good in other things, and they’re also going to need some training, some time with the actual tool which you want them to use. So they’re going to need both of these things in order to use AI in a meaningful way, and of course, really important to also give them the time and space. What I often hear from people is that, oh yeah, we have this AI tool now. I tried it, but you know what? Using the AI tool actually takes me way longer than doing this the way I’ve done it for the last 20 years. So I’ve stopped using the AI tool, which, of course, as we all know, if we took the time to learn how to do it with AI. That will probably result in you doing it quicker and more efficiently. But I also understand people who, again, haven’t been given the skills, haven’t been given the why and the context that they might then just very quietly drop it. We’ve already mentioned how important it is to bring people along. The workforce not adopting AI or even quietly sabotaging it, if you will, is one of the biggest reasons why AI doesn’t stick. So you definitely don’t want that to be your workforce. So know who those people are and give them the right skills to be able to succeed with AI. Yeah.

Catherine McDonald  22:58

No. Really, really good points. And your points on, I think what I’m drawing out of that is also the role of the leaders or managers in in this case for of every team because training training is great. We all need training. We all need a step by step guide. We need to be shown things for sure. But the role of the manager or leader is to continue on the informal coaching and training long after the workshop ends, and to build that into how they do their work every single week through whether it’s one to ones or or team meetings. But if we don’t bring AI in as on top of everything else that we do in our one to ones and our our weekly meetings, we are not going to see it succeed, or it’s just going to drop off, like you say. So, there’s there’s two things in there. It’s one making sure the fundamentals are in place in terms of the the managers understanding their role in terms of training and mentoring and coaching, and then the other is the organizational approach to AI, making sure the managers are clear on that, so they know what they are actually training, coaching, and teaching people in. So quite a lot to that, and quite a lot for managers to get their heads around.

Dajana Achelpohl   24:08

There is quite a bit, and what you really want to avoid is what I often see. Oh yeah, we’ve done all an AI. We’ve all done an AI training. It’s one of these self-paced, pre-recorded things that everybody just quickly clicks through and retains zero from. So you really want to find a way to make this very tangible for people, as you said, be it in one-on-ones, in team meetings. You want to give people an opportunity to really showcase. This is how I’m using it. This is what’s working. That’s this is what’s not working, etc.

Catherine McDonald  24:38

Excellent, great points. Okay, so we’re on to the what if.

Dajana Achelpohl   24:41

We’re on to the what if. So AI obviously is wonderful, but we all know that it makes mistakes, right? And that there is risk associated with AI. And I can’t stress this enough: AI will make mistakes. So it’s not a question of if it makes mistakes; it’s a question of when. So we need to be prepared for that. So first of all, we need to consider what could go wrong and what needs to be in place. What needs to be in place, both from a legal requirement in the EU. Now, for example, we do have the EU AI Act in not quite full force, but in some force, even more since August this year, so there might actually be legal requirements in terms of what needs to be in place if you’re using AI. But also think about what potential damage this can do to your brand, this can do to your customer relationships, etc. So you should have a really clear understanding of what can go wrong, and what are we going to be doing in that scenario? The concept of humans in the loop is, of course, very big in AI. We do not want to outsource the the last sort of decision, the last sign of to AIs in many ways. How is our work going to look in future? We might be saving time here, but on the other hand, we now need people to have time to check AI outputs. We need to teach them how to do that well and what to look out for. Another important thing in that what if scenario with the risk and guardrails is that I try to look at the data that’s there. Data, obviously, the oil that powers the AI engine, your data needs to be good, usable data in order to see good results from AI. But again, you need to ask the question there: What data can we use? What data can’t we use? Be it because we haven’t collected it for this particular purpose, be it that the data is not fit for this purpose, be it that we do not want to use the data for this purpose, or also have organizations that are very clear on what data to use and what data not to use. So that’s a whole big thing. Then, of course, we need to talk about things like hallucinations. We all know AI actually makes up things that look very credible and read wonderfully, but are just not correct. How are we safeguarding against that? Again, how are we skilling our team? So I think there’s always an important case here to be made for having clear governance in place for AI, for having very clear AI use policies, I work with many organizations where there is an AI use policy, but it’s a 48 page document that nobody has ever looked at, and people are still very unclear. I think it’s important to have an AI use policy. I don’t think that it prevents people from using AI. I actually think it gives them the confidence that they know what to use AI for and what not. So it gives them that permission. So when we ask “What if? What we’re really asking about: Do we have all those checks and balances and those guardrails in place? Yeah.

Catherine McDonald  27:58

Excellent. Okay. Yeah. Really. Really important points that I would imagine not a lot of people even know about. So yeah,

Dajana Achelpohl   28:06

no, there’s there’s a lot of things to to get people worried, things they hadn’t even worried about.

Catherine McDonald  28:12

Yeah, but again, aren’t you better knowing? Aren’t you better knowing about the need for the policies or the legislation that’s out there? You’re better knowing, so it doesn’t have

Dajana Achelpohl   28:22

to be as complicated as people think, as well.

Catherine McDonald  28:24

No, not to put us off, but just to make us let us do our best right now to make sure we we embrace AI safely. Yeah, and then the final question

Dajana Achelpohl   28:34

is: Then what? Then what? This is not one’s exercise, and you’re done, right? It’s not a one and done. So, in order to make sure that AI adoption sticks, you are going to have to consider what happens after that pilot. First of all, coming back to these organizations I work with that have all those pilots running. First of all, let’s put a very clear timeline on this pilot when we will stop this pilot if it’s not working. But be very clear what happens after. How do we keep people engaged? How do we encourage AI use on an ongoing basis? How do we keep people learning? These tools evolve constantly. There’s new abilities all the time, so this needs to be ongoing. How do we receive feedback from our teams on how AI is working. What’s not working? How do we incorporate that into our processes, into our tools? Do we reinforce this from a leadership point of view? AI needs to be something we talk about regularly. It can’t just be a once-off exercise and then it’s all done. So yeah, there are the five questions. Quite a few things there to think through, but this is quite comprehensive, right? So once you’ve worked through this, you’re going to have a really good idea where you are at and where you should focus.

Catherine McDonald  29:51

I love that simple structure. I just think for a complex topic like AI that’s so new to us all, that’s exactly what we need. Structure to help us deal with it. So amazing, Diana. Who do you think should ask these questions, and to whom?

Dajana Achelpohl   30:08

Yeah. So I think it obviously depends on on the organization and how they are set up. I often see these organizations asking these questions at functional level, and then it boils up further, right? There’s some elements in there which you probably need to clear at the overall organization level. So if we’re talking about governance, AI use policies, etc. you want that to be the same across the organization because you don’t want to create silos there. Other questions, however, why are we using this? What does our work look like, etc. I think you can ask that at the team level. I see a lot of function leaders that are asking those questions, and ultimately, I think you can nearly use them at every level of the organization.

Catherine McDonald  30:58

Good. Okay. Great. I found that really, really helpful. I really believe our listeners will as well. So, Diana, if people want to reach out and talk to you, or just follow you on LinkedIn, or find out more information about the work you do, how will they find you?

Dajana Achelpohl   31:15

Yeah, absolutely. So, I’m on LinkedIn, Diana Achelpool. Not the easiest name. You might have to take note of that, and you can reach me at diana@aichangemaker.com.

Catherine McDonald  31:26

Perfect, and it’s D A J A N A.

Dajana Achelpohl   31:29

Yes, yes. I’m sorry, that’s what it is.

Catherine McDonald  31:32

No, perfect. Okay, so Diana, it has been an absolute pleasure. I’m really glad we got the opportunity as well to chat, and thank you so much for sharing your wisdom, and we will definitely stay on Twitch and LinkedIn ourselves. And I hope you enjoy the rest of your day. And we’ll say goodbye to our listeners. And thank you for listening or watching. And we’ll see you again on the next episode of the Lean Solutions podcast. Bye for now.