Nov. 4, 2024

Athena Peppes: The three P's of AI Success

Athena Peppes: The three P's of AI Success
Athena Peppes: The three P's of AI Success
AI Pathfinder for Private Equity Podcast
Athena Peppes: The three P's of AI Success

Summary In this episode of the AI Pathfinder for Private Equity podcast, host Steve Budd engages with Athena Peppers, a thought leadership expert, to discuss the critical components of successful AI mobilisation in private equity. They explore the current state of AI adoption, the importance of understanding its impact on various industries, and the challenges organisations face in implementing AI strategies. Athena emphasises the need for a clear problem definition, the role of people and cu...

Summary
In this episode of the AI Pathfinder for Private Equity podcast, host Steve Budd engages with Athena Peppers, a thought leadership expert, to discuss the critical components of successful AI mobilisation in private equity. They explore the current state of AI adoption, the importance of understanding its impact on various industries, and the challenges organisations face in implementing AI strategies. Athena emphasises the need for a clear problem definition, the role of people and culture in AI adoption, and the significance of partnerships in navigating the complexities of AI technology. The conversation also highlights the importance of data readiness and leadership buy-in, as well as the common pitfalls organisations encounter during transformation efforts. Athena shares insights on learning from industry examples and the value of diverse perspectives in understanding AI's potential.

Takeaways

  • The adoption curve for AI varies by industry and company size
  • AI transformation is more complex than past IT transformations
  • Focus on problem definition to guide AI initiatives
  • People and culture are critical for successful AI adoption
  • Partnerships can help organisations navigate AI challenges
  • Data readiness is crucial for effective AI implementation
  • Learning from industry examples can inform best practices

Links
Athena LinkedIn profile
Beacon Thought Leadership

AI Pathfinder Private Equity AI Strategy Meet-ups:

https://www.aipathfinder.co/private-equity-ai-events

Have a question or would like to suggest improvements, please contact steve@aipathfinder.co

SPEAKER_00

Hello and welcome to the AI Pathfinder for Private Equity podcast. I'm on a mission to lead private equity from AI bewilderment to AI empowerment, and I'm doing that by building a dedicated network that will piece by piece create a playbook for successful AI adoption across private equity operations and the deal life cycle. I'm meeting and having very interesting conversations with people within PE and across the AI value chain. I'm really pleased to bring some of those to life through this podcast. Today we're going to talk about the building blocks for successful AI mobilization, which is the theme for my next private equity AI strategy breakfast briefing in London on the 3rd of December. Now, a recent report by RAN found that 84% of business leaders believe AI will have a significant impact on their business, but the majority aren't ready, leading to as many as 80% of AI projects failing, twice the failure rate of other IT projects. So how might private equity reverse this trend? Well, joining me today to talk about this is Thought Leadership Expert Athena Peppers, Technology Trends Advisor and Director of Beacon, an Accenture alumni. Hello, Athena. Thanks so much for joining me.

SPEAKER_01

Hi, Steve, and thank you so much for inviting me. Pleasure to be with you this morning.

SPEAKER_00

Thank you. Well look, first things first, let's get a little bit more context on you, Athena. So I'd like perhaps if you can, briefly tell me a bit about yourself and what you're doing now and where AI fits in.

SPEAKER_01

Sure. So a bit about myself. I'm an economist at heart, but I've spent quite a lot of my career focused on researching and developing thought leadership for uh business topics and particularly focused on the future of business. And obviously AI is an important part of that at the moment. So as uh one of the directors at at Tobacon, uh we help companies shape their thought leadership. And you mentioned AI empowerment at the beginning when introducing this, so that's kind of quite important because by doing some original research and developing your own thinking, um, you know, your own thought leadership on this topic, um you know, it's kind of an essential element of AI empowerment as well. And that's the that's what we are helping companies with at the moment.

SPEAKER_00

Okay, so how do you see things have changed in terms of well, certainly the thought leadership? What what are you being asked to to do in in in the relate in relation to to AI more specifically? Are there this sort of concerns about the future state of of certain industries, for example?

SPEAKER_01

Yeah. So I remember working on a report on AI back in 2016, actually, uh, and that was focused on understanding its economic potential and impact on productivity. So, in many ways, you know, in the in the field is obviously been around for a lot longer than that. Um but I think what's changed over the course of the last few years, and particularly the last two years, is that it's kind of come more into the the public consciousness. Um, so there's a lot more interest in understanding what AI is, what the impact it can have could be. So, you know, obviously the the launch of Chat GPT and generative AI becoming more commonplace is what changed that for coming up to the two-year anniversary of the of the launch in in November. So we are seeing companies kind of you know more interested in the technology. Um I think one of the challenges that I'm seeing is that uh it's quite easy to talk about perhaps the cost savings, you know, the and the the easy wins. Um what's a bit harder um is to imagine how it can actually change the future of your industry. How can it create new growth? How can it augment a lot of what you do? Um, you know, as an organization, for you individually, for your teams. So I think that's how you asked about you know the evolution, right? So those are the two dimensions. One is definitely a lot more interest over time uh from from all sorts of companies. Uh, and the other is uh that that shift and that kind of organizations being keen to go beyond just the cost element to the you know new growth element um that AI can unlock.

SPEAKER_00

Yeah, and I and I assume that that some of this is driven by concern as well, um, around the validity of potentially some business models uh in the future as well.

SPEAKER_01

Yeah, I mean, um absolutely, and the concerns can be quite broad broad uh range. You know, there's concerns around the impact on uh on jobs and how all the concerns around kind of how do you help your people use the technology, for example, effectively. You talked about the RAND research that shows that the failure rate of AI is um for AI projects is twice as high as that of just, you know, uh IT projects. And part of the reason for that is people not being able to actually use the technology. So you roll this out, it's seen as a technology transformation, but actually the people transformation is quite important uh element of it. Um there are concerns around um data privacy and you know, the data that you share on these models, how will they be used in the future? Um, are you, in a sense, giving away interesting information that might mean in the future you're out of business? Essentially, like what's happening with some of the creative industries, right? Where, you know, those tools, the recordings of podcasts, for example, are being used to train uh, you know, new tools that can create podcasts. I don't know if you've seen that, the the Google LM uh tool, where you can basically give it an article and it converts into a podcast. So, you know, I think that can apply to any industry, right? There's a concern that, well, if we share information about a competitive advantage while using these tools, um, because sometimes people might not realize that they're giving away that information as well. Does that mean that in the future um we'll have, you know, we're giving away some of that information to technology companies potentially? Um and there are, I guess, ways around that, like developing your own models and things like that, or customizing them, putting support um systems for your data around that, but we can go into that a bit later.

SPEAKER_00

Yeah, I think, you know, well, perhaps summarising eyes wide open, really, and um and thought leadership is helping to, you know, to to fill any any gaps there in terms of what what the future may hold. Well, let's you mentioned certainly people there, and um and if we just you know go into mo what i AI mobilization is now you know mobilization is is a big factor when we're talking about any form of trans transformation, particularly digital transformation, but as as you as we mentioned above, is the uh the stat around potential failure rates and so getting those foundations, all the components uh in place, not necessarily before you do anything, but it's an important for us to recognise what those things are. Um, you know, you know, question for you from your perspective, what what is AI mobilization and and why is it important for business to think of AI as as more you know more than just uh this is a sideline tech project?

SPEAKER_01

So um that's a very good question because there has there have been already instances of companies that are piloting, experimenting, rolling out, you know, these um these tools across the the company. You know, you get debates around oh, but why are we not seeing some of those productivity benefits yet? You know, at what point will it reach, you know, the the point where we kind of start seeing that return on the investment as well. Um and I think that, you know, a lot of different factors come into it. One is time, right? Uh perhaps we expect things to happen a lot sooner or to see results a lot sooner than that than we should. Um with projects around technology transformation, particularly uh AI, as you mentioned. It takes time for it to feed through. So at the point where it's adopted throughout the organization and can start having that impact. And we're just not at that stage yet, you know, if you think of an adoption curve. And that adoption curve might vary depending on the industry and also the size of the companies you're talking about. Uh, of course, when you're looking at, say, the tech sector, they're a lot more mature on that adoption curve. But if you're looking at other sectors, there was some research actually done by uh McKinsey looking at that curve. Manufacturing is a bit further behind. Uh, they're more at the experimentation and pilot stage, consumer goods a bit further ahead. So each one is at a different stage. The other thing that comes into it is the size of the company as well. The budgets kind of allocated uh will differ across the size of the different size companies. Um, a lot of the research that I've seen and the the the thought leadership on this topic has been focused on very large companies. Okay. So of course you might see them kind of being a bit further on that adoption curve. But when you look at um research around, say medium-sized companies, there's definitely more of a gap in in that space. And you know, initial um research uh actually shows that not as many of them are adopting it. So why why am I saying those two things? Because they're kind of part of seeing the benefits of a technology that economists call general purpose technology, like AI, like you know, electricity, things like that, is that it it takes time for the impact to feed through and to to kind of see the results in in the numbers and in the way that we we work and we function. Um and two years is probably not not enough time given these tools are relatively new.

SPEAKER_00

Yeah, and there's um cultural thing here that's um that's it that's obviously quite important to uh any sort of transformation, but it feels like there's you know there can be appetite um with emerging technology like AI to make a move and that can go come from you know the top, but rather than thinking about mobilization, it's almost a knee knee jerk, and so we're gonna do something. And there's many proof of concepts that have been built, but then you get to a point where it's all pulled back because the mobilization piece has hasn't happened, and that could be down to you know, it could be down to the people, the capabilities you've got, the uh the infrastructure, and and obviously they're really not having a a joined up strategy. There's lots of reasons why people are finding it more difficult than maybe they expected to, uh, with you know a a magic technology. So yeah, it's I mean it'd be good to think back at other sort of digital transformations and and you know, is mobilization better understood in other areas of the business, um, or is it is it just that there's a greater demand to do something here so people are trying to move faster and hence why there's more failure?

SPEAKER_01

I mean that that uh my my research uh kind of uh uh dimension is kicking in, and I'm thinking that that's something actually that would be interesting to to look at. But I think you already touched on some areas, which is that AI transformation can be quite different to past waves of IT transformation that that we've seen. And obviously, as humans, we all expect it to be similar to what we've seen before, but actually it can be a bit more complex, partly because the technology is more complex and it might take more time for people to understand how to use it. The other thing is that, you know, looking back, we feel like maybe things happen faster, but actually it took a long time for these things to to feed through, you know, and taking like a bit of a different, you know, a kind of broader perspective can help you think about that. So even things that we take for granted, like electricity and the internet, are not uh commonplace in some parts of the world, right? So there's always that divide like different types of divides um that that come into it. One thing that that helps me think about you know a AI mobilization, what's what kind of would drive success is what I call the three Ps. It's just a good way to to think about you know what are the enabling factors. The three P's are problem people and partners. So the problem, um the problem is really the idea that you you should be focused on uh when you're when you're thinking about uh doing pilots or how to roll out AI, you need to be very clear about why you're doing it. I think there's a risk that because of the hype, because of the you know, the public interest, oh, we must do something on AI. What should we do? Let's just do this. And then the stakeholders are kind of underwhelmed by the results because they weren't necessarily focused on solving a particular problem for the business. That problem might be a today problem or a future problem. Today problem might be more kind of, you know, the cost aspect that we mentioned. The future problem might be, well, what you know, how can I develop this market or how can I think about new business models? The second element is around people. And that's where kind of bringing them on board is important. So we talk a lot about skills and upskilling. Of course, that's important, right? Being able to use those tools, understanding how they how they work, what you can and can't use them for. But also there's you you mentioned the word culture before. There's an element of that and bringing people along with you. Um, I think that's a a bigger challenge perhaps with AI, because people fear it to some extent, right? There is that kind of like, well, but if I use this tool, could it take my job? And therefore, like there's a vested interest in me in me using sorry, conflict of interest in me using it because it might actually replace me. In fact, you know, it it could make it more productive. But I think that kind of that um psychological and cultural piece as well definitely comes into bringing people along with you and successfully mobilizing um for AI. And the third thing is is partners, right? Sometimes it can also appear as quite an overwhelming thing to do, and of course it is, but can you, you know, can you look to other uh organizations, whether in your industry um or you know, different sets of partners that can help you create an ecosystem that delivers that project? Perhaps you're not sure where to start with your data, right? We haven't talked about data yet, and that's an important element. Could you work with a data audit company that helps you understand what kind of data sets you have, what's proprietary to you, um, how you can, you know, use it in a way that also fits with all the regulations and requirements that are coming our way. Um so you don't have you don't need to have all that capability in in-house. You can lean on kind of partners in the broader ecosystem to help you do that. So, yeah, problem people and partners are kind of three ways to think about the the enabling factors and uh you know how to successfully mobilize for AI. I mean, the other thing that's unique to every company is of course the data that they hold, right? So you mentioned like what are kind of some of the practical steps to to to get started. I think the the problem definition is always the first thing. Um I know you talked about leadership buy-in, but ultimately uh in my experience at least, it tends to come from leadership buy-in, like kick starting those projects. They're very hard to do, you know, grass grassroots. You're not gonna get someone doing that. You might get people using the tools and you not knowing about it at the leadership level and how it affects, you know, and therefore not capture the the benefits and your KPIs, for example, because you're not aware of what's happening. But the way it happens most times is that there is, at least from the function that starts it, some leadership buy-in, some budget. Now, whether that is the case across the board, you know, whether you're siloed, of course, that's part of the success, uh, you know, of the success factors that come into it. If you're if you don't have the buy-in in terms of the rest of the organization, then they might not see the benefits or be willing to do so. So that's why I haven't kind of focused on on leadership buy-in too much, because I think that's where those projects tend to kick off. Someone will say, we think we should have a pilot around this, you know, here's some budget, see what you can do with it, uh, and you know, wow me with the results. And that's where things can unravel because it's a bit more complicated than that, you know, it's not a discrete kind of um uh tool. So um once you have that though, once you you have got started and you've got that buy-in, then understanding what proprietary data you have is quite important. And every company will have some proprietary data. Um now what's what's quite interesting as well is the kind of structured versus unstructured data dimension of the business. So loads of companies tend to use mainly structured data that they've got, but the real kind of insights can come from unstructured data that you have. So I don't know, in a P context, that might be uh conversations you have, you know, with people, or that you have as a team, you know, brainstorming about some ideas that are not, you know, captured somewhere or codified in a spreadsheet. That that would be unstructured data that you can capture, turn into insights. And the good thing is that Genitive AI in particular can use unstructured data quite well. So you don't even, you know, it can work with that. You don't you don't need to convert it into something else essentially. It learns from that unstructured data. So we've got the leadership buy-in, then you've could kind of kind of uh do your data uh leadership buy-in, problem definition, uh then data uh audit and kind of strategy around that. Um and then once you've uh kind of assuming uh assuming the the pilot's done well and you're looking to roll that out into other functions and scale it, then what you need is a lot of the people element, the culture element, making sure everything's integrated, you're taking people along in the journey with you, as we've discussed, tracking the progress, that's another challenge that a lot of companies have in the sense of some of the benefits are quite hard to capture. Uh if if it takes me 10 hours to write a piece of code instead of 30 hours, I can track that. If I come up with a new idea, how how is that kind of captured in the KPIs and the metrics that you're following? I think that's something that organizations in general struggle with, capturing some of the intangibles that are actually quite um beneficial. It's not just an AI problem, it's a more general problem when it comes to being innovative and creative.

SPEAKER_00

Moving on to the challenges, the common challenges and mistakes that that companies make. Not necessarily just mobilizing for AI, but mobilizing for other transformations that that have gone in the past. What are your thoughts on the watch outs there?

SPEAKER_01

If we're if we're to take a broader lens in terms of transformation, business transformation more broadly, um the research kind of consistently finds that seven or eight out of ten transformations fail, and the primary reason is of those that fail is is not taking the people along with you. Whether that's in terms of willingness to adopt and use the tools, I could give you something that's amazing and invest a lot of money in it. If you're not going to use it, then that money is completely wasted. So how do you make sure A, that they can use it, right? They know how to, you know, that's the kind of skills element. But also how do you kind of showcase the value that it can bring to their day-to-day and therefore actually adopt it, you know, and make use of it so that you can actually start seeing results that that are captured by the by the business metrics more broadly. That's what you know the business transformation was a topic I I researched quite a bit. And that's that's what the literature kind of shows is the biggest uh the biggest factor that that comes into it, people. The other thing to bear in mind, and you you touched on this a few minutes ago, is that every organization is so different, right? So yes, there are some best practices around identifying your data and you know, buy-in and things like that. But ultimately, the route you'll take also depends on what that cost-benefit analysis looks for your own company, right? So uh and that's why you can end up with very different models because you can look to what others are doing, and that's great for inspiration and best practices, but ultimately you need to find the balance that works for your own, you know, for your own company's um organization or organization.

SPEAKER_00

So yeah, and you've you've talked through some, you know, other points around the challenges is you know the it t takes time, so you know, that there might be an overestimation in terms of what the short-term gains may be. This is definitely a long, a long uh game that we're playing here. And the lack of data data readiness is that you haven't put the prep in. Um they haven't sort of been able to uh find where your you know your unique data is and it's or it may be in a many different forms, and you've just got to find a way of of of uh of putting that together. But absolutely, I think that there's a lot of opportunity, it's just about eyes wide open to those those challenges and the the need for for a suitable mobilization. Just just into the final sort of very quick questions, really. Is there anyone out there that's that's doing this well? You know, are there good examples in your opinion and that's that's doing this right, that you feel have taken that? You've said that bigger businesses, so that may be um you know, that may be a benchmark for us. Um yeah, so so that we can look at in terms of best practice.

SPEAKER_01

I'm not sure I would pick any one company because as you said, it's still early days, and I do get the sense that the field is dominated. The field in terms of like, you know, the the case studies that you hear about, right? And there's publicly a lot of publicly information, uh publicly available information around, they tend to be the large companies, the tech companies. So, you know, for me, uh the important thing is to to kind of check that bias as well and think, okay, that's what they're doing. They seem to be leading. And yes, as I said before, we can look to them for best practices, but what works for my organization, my industry, um, that that will be quite different. And, you know, so I think that's quite important that the the narrative that we get is quite heavily influenced by those two, you know, kind of areas. So it's quite important to maybe maybe that's a gap in the market, right? In particular for for private equity, what's the state of play in in private equity, that there's perhaps a need to to kind of explore that, research that, understand that better than it has been so far. Um so I think that's that's definitely uh quite important. There's a gap there.

SPEAKER_00

Yeah, definitely the mood music that I'm picking up um with a sort of cohort cohort of of PEs that they're looking at their peers. What are their peers doing? And there might be a press release that's released, but you know, it doesn't really give you much detail. It's just you know, maybe, oh yeah, we're doing stuff with AI. Um But there's there's a sense of well, we might not move until we see our peers moving. Yeah. But that could be too late. So it's I think it's yeah, it's it's really important that um that time is taken to to review what those opportunities might be. And that could be at the very, very early stages, just getting a a group together that could form your task force and bring that group up to a certain standard where they can start uh investigate further what the opportunities for the business or the firm firm may be. But yeah, I think just waiting may not be the right answer.

SPEAKER_01

There's plenty of examples of companies that perhaps saw a trend, saw something, didn't act on it, and then, you know, I can give you some examples that I've been following in the news. So I think De Beers is quite interesting. I know it's not necessarily in the private equity space, but just more broadly in terms of companies, right? They they talked quite a lot um in their past releases about the fact that lab growing diamonds, which is something that a few years ago they didn't think was, you know, important to them, is actually wiping out uh quite a lot of gains for them. So it's kind of they knew the trend was there. They could see that this is, you know, there's a new technology that enables companies to manufacture diamonds, but they thought, oh, this technology is, you know, not going to affect us. And it has, you know, it has affected their bottom line. Um, the same happened in in healthcare, where the companies have been investing for a long time in weight loss drugs. And, you know, it's not like we didn't know about this. The the technology was being developed. But companies like Weight Watchers have seen their share price go to like one dollar now from a hundred dollars because their you know their business model has been completely disrupted by by innovation and new technology that was on the horizon but they didn't act on. So I agree with you that it is important to, you know, not wait until it's too late for sure.

SPEAKER_00

Last question. Um, because we're definitely come coming to time now, but is there a single source of truth when it comes to keeping track on AI for you?

SPEAKER_01

Um I'm against that because I think that you can get blindsided by just following one person as the single source of truth. So I I follow a lot of different people as a way to understand different things. Um I know at the breakfast that we were uh that that you had a couple of months ago, the AI Pathfinder, the last one you mentioned, Ethan Mollick, who's great. Uh, but there's there's loads of other people that kind of take very different perspectives, global perspectives, you know, more um kind of concerns around ethics, regulation, data. So yeah, I there's not just one person. I think it's important to follow different people so that you can see what's happening across different lenses.

SPEAKER_00

Yeah, and I suppose in your your role anyway, you're you're having to dig into lots of different sources to find a balanced perspective.

SPEAKER_01

Yeah, exactly. And kind of the interesting nuggets can come from questioning what's you know commonplace knowledge everywhere as well. So, yeah, absolutely.

SPEAKER_00

Yeah, look, well, this is part of the process of myself learning and our listeners learning is talking to to different people um that have different skill sets, come from different parts of the organization. I think it's really fascinating to see how their you know their what their AI journey is, and it's it's it's different. There's I I couldn't I wouldn't be able to categorize people yet and who's um leaning into it or or finding it um maybe stressful. I really enjoyed the the conversation today with you, Athena. Thank you so much for your time. I assume that you're happy for people to connect with you on LinkedIn and um and also to check out uh Beacon at Beacon Thoughtleadership.com. But for now, thank you very much for listening, everyone, and um, we'll see you next time.