208. Leadership in the Age of AI

We love hearing from our readers and listeners. So if you have questions about the content or working with us, just get in touch on [email protected]

Say hi to Sophia on Twitter and follow her on LinkedIn.

Following us on YouTube, Facebook, Instagram and TikTok will make you smarter. 

Episode Transcript

Sophia Matveeva (00:00.142)
Do you think AI is currently overhyped or underhyped? That's a very good question because that's what everyone is talking about.

Sophia Matveeva (00:11.79)
Welcome to the Tech for our Techies podcast. I'm your host, tech entrepreneur, executive coach and Chicago Booth MBA, Sophia Matveeva. My aim here is to help you have a great career in the digital age. In a time when even your coffee shop has an app, you simply have to speak tech. On this podcast, I share core technology concepts, help you relate them to business outcomes,

And most importantly, share practical advice on what you can do to become a digital leader today. If you want to have a great career in the digital age, this podcast is for you.

Hello, smart people. How are you today? In this episode, you are going to learn some sensible approaches to leadership in the age of AI. As I'm sure you've noticed, AI is a really hot topic right now. The day that this podcast episode comes out, in video,

The chip maker that can train and run generative AI models like Chagy BT has surpassed Microsoft and Apple to become the most valuable publicly listed company in the entire world. And guess how much it's worth? It is now worth $3 .3 trillion. my. So AI is clearly making some people a lot of money. But when a technology receives this much hype,

then people start losing their minds and put the technology on a pedestal. And this can have really negative consequences. So here's an analogy for you. Do you remember when you really last fancied somebody like you just thought that they were the sexiest person in the world? Did you see that person objectively? Absolutely not. So I think we're kind of in the same place with AI right now. This is why for this episode,

I invited an expert on both leadership and technology to discuss how business leaders need to approach AI. You are going to hear from David Decramer and he is the Dean of the Diyamorim Mckim School of Business at Northeastern University. And he's also the founder and director of the Center on AI Technology for Humankind at the National University of Singapore Business School. David is known for studying the role of leadership in the context of AI adoption and

Sophia Matveeva (02:30.958)
digital transformation. He has way more accolades than I can list in this episode, but he's very smart, very successful, taught at Cambridge and so on and so forth. He's the author of several books and his latest is called the AI Savvy Leader, Nine Ways to Take Back Control and Make AI Work. So let's welcome David. Welcome to the Tech for the Techies podcast. Thank you, Sophia, for having me. Happy to be here. So my first question to you.

is do you think AI is currently overhyped or underhyped? That's a very good question because that's what everyone is talking about. I definitely do think if we look at what is happening within business in organizations, AI is overhyped. And the reason why I'm saying this is, as you know, most of the AI, the amazing, wonderful stuff that we see that gurus are promoting that we read about in the magazines has been developed in a lab.

And a lab is a context where basically you don't have any stakeholders to take into account. You're not going to harm humans right away. But once you bring it into business, you have a range of stakeholders to take into account. You need to take care of their interests. So risk management goes up exponentially once you move it into that context. And what that implies is that the AI that we're all going crazy about and that we see what is the

potentially possible will not be possible in business today because your risk management will make that you're using not the latest version of AI or the most hyped version of AI because you can't. It's not tested. We don't know the consequences. Your risk management will downplay that. And that's what I always say to companies. Most of the AI today is still supervised machine learning, which means of course there's a human in the loop.

Dreaming about artificial general intelligence, please be my guest, no problem. But it's not going to happen in the next five years in your company. And especially not because most companies are not even well prepared. If you look at the numbers, we see that most digital transformation projects, AI adoption projects are failing because they can't upscale. And one of the reasons there is the lack of business direction, business leadership in saying what is it that we want to do with AI?

Sophia Matveeva (04:55.758)
for which organizational purpose. And that's also the reason why I wrote my book, of course. So I was actually going to bring that up because in your book, you said when it comes to AI driven transformation, leaders are not leading. And for the people watching this, this is the book, the AI savvy leader. So what do you mean that business leaders are not leading when it comes to AI transformation? Yeah, well, it's basically the observation that guided the writing of this book.

So in my work as a consultant, as a business school professor, and now also as the Dean of a business school, when I talk to companies, one of the first things that I see is that most of their budgets up to 70, 80 % of the budgets reserved for technology and transformation projects will go to the adoption, the requirement of the technology. And that's where business leaders leave it. Because as you can imagine, most executives today, when they were educated, there were no topics.

like AI on the curriculum, there were no topics like sustainability. They weren't thinking about these things. So the problem that most business leaders today are facing is they have to get to know this new technology, which is basically a tool to use while at the same time adapting to it. It's a conundrum they're faced with. And that's a very difficult one. So how can you learn about it while you don't even know it? And that of course induces fear and that's...

periods enhanced further by a sense of urgency because just read any magazine, any business magazine or newspaper today, what is the, the, the major statement being made? The biggest risk today in business is not using AI. So of course there's a fear of missing out and a lot of business people feel inadequate. So what happens, and this is what I call in my book, the myth of technology driving technology transformation, business leaders start believing in that myth.

It's only tech experts who know what to do with this technology and make value and create value for our business. But this is where it goes wrong. Because as you know as well, technology again, AI is a tool at the moment. It's a tool that we use hopefully to create value, value for all our stakeholders. Now, how do you create value? By first of all, knowing the relevant business questions. You need to know what is it that you want to create as an organization? Why are you in business?

Sophia Matveeva (07:21.71)
Who are your stakeholders? How are you going to work with this? These are all business questions. So now the question becomes, are tech expert business experts? No. So that's why I said most leaders sideline because they're not AI savvy enough. And once they have that AI comprehension, they will be able to develop a narrative where they can say, these are the business questions. But now I have a narrative where I can tell a tech expert, I think AI...

Data management analytics could be used here to address that question. And at that point you delegate because the tech experts will drive the execution. What we see now is tech and AI enters the organization. We've done the budget part and we already delegate, which is too early. It's interesting that you're seeing this as well, because this is literally the reason why I set up TechFund on TechAid because...

No, when I came out of business school, so I was at Chicago Booth, and that's where I started my first tech company. But before I worked in private equity, so very, very different field. And that's when I really saw that there exists divide, you know, there are the computer science students, and then there are the business students. And then, you know, as you get into the professional world, you have the business leaders, and you have the technical leaders. And in a lot of cases, the two don't really know how to speak to each other.

And there's also kind of fear. There's definitely fear from the business leaders that, you know, the tech people are doing something very complicated. I won't be able to understand it. And in order for me to even talk to them, I basically have to take coding course. And of course, lots of people, most people don't want to take a coding course. And so then they kind of take themselves out of the conversation. But then you're right. You know, when you have investors breathing down your neck saying, okay, what are you doing with AI?

You've got to have an answer. So you end up in this awful situation, but I'm curious what your view is. What is it that a business professional needs to know about AI to be an effective leader, to know when to delegate, because there's that fine line, you know, they need to learn something new, but they also don't need to learn everything. Exactly. It's a very good point. And that refers to the AI server leader in the title of my book.

Sophia Matveeva (09:45.198)
And what is that sevenness? That's an important question as a Dean of a business school myself. Of course, I'm asking this question every day almost. Where do we need to guide our students? Which direction? And I always say, look, as a business school student, you don't need to know which large language models there will be in three years or five years time. That's what computer scientists need to know. And they won't even know either because things are going so fast.

But what you need to know, and so I implemented an AI strategy in our business school where we start from the assumptions. AI is a tool. It's a tool to create value and preferably holistic value going from promoting efficiency and productivity to wellbeing. To innovation. It's different dimensions. And for example, regenerative AI, it's not simply like following it. You're not following AI. You're still leading AI. So that means.

It's not just using content. It's really about the transition from content to knowledge. Same thing in business. So that's what I expect from my students that they can use AI. I'm fine. Use it in the, in the classroom, but you're not being evaluated anymore based on what you generate. You're being evaluated what you do with that generate generated content. And this is same for business leaders. They need to understand today. The leadership behaviors that were relevant in the past, like being visionary and powering.

creating trust, making sure communication is clear. There's a feedback circle cycle. It's still relevant today, but with this difference, there's a new type of coworker which uses a tool. So you need to leverage the human abilities and at the same time have enough comprehension of AI so that you can augment that human intelligence. And that's where your company is going to perform better, faster and be more competitive. So going back to your question then, so what is it that they need to learn?

This goes of course, beyond mere coding experiences. This is literally about what is human intelligence about? What is artificial intelligence about? What are the weaknesses and strengths? So in my classes, and I do this for boards as well, because they're even in more need to understand AI is, this is what is uniquely human. This is what AI will not do. So just explaining what is supervised, unsupervised. It's statistics. It's generating it's...

Sophia Matveeva (12:10.382)
It's probability, it's predictive, what do these things mean? And then you can compare it to what does a human do? Because actually we need to stop with the narrative of comparing artificial intelligence so explicitly with human intelligence. It's comparing apples and oranges. They are different. But if you can get that, then you know how to best use your people. And at the same time, use AI for the routine tasks. Use AI.

for certain procedures that will help your people to work on their strengths. Once you start understanding that, you can develop a narrative. And that's actually the AI7 leader today. Basically a narrator who brings down the silos that you refer to, the tech experts and the business experts, because they don't communicate. They don't communicate with students and they don't communicate in the business world. So that's... But don't you think that...

what you're talking about really refers to the white collar profession, because yes, you know, for example, you could have, I don't know, a marketer. So for example, my social media team has been using AI tools to, you know, chop up the YouTube videos of the interviews that I do on the Techful and Techies channel, for example. And that has been a huge time saver, which is wonderful because then they can do far more interesting and creative things than chopping up videos. So,

as an employer in a professional services firm, I'm very happy. However, if you look at say, a card, which is a British supermarket, and you look at their warehouses, that's basically what they're kind of like Amazon warehouses where they are really operated by robots and they're operated by AI. And in that case, there are blue collar workers that are basically not doing those jobs anymore. So.

Isn't it the case that in some cases, so in the blue collar workspace, AI is replacing jobs, but with white collar workers, it is a work enhancer. What do you think of that Steve? I love the examples that you're providing in your industry, which is a creative industry. Yes. AI can help you focus on the more creative aspects. What am I going to do? As I said, with the content generated, I don't have to cut up the clips. I don't have to do all that.

Sophia Matveeva (14:35.95)
Teen work, which is mechanistic, basically. I don't have to reduce myself to this anymore. I can use it as a tool and run with the content. Great. And that's what we all like to see. But of course you have a point there. Not all our jobs in the world are creative jobs. So what about many of these other jobs? And then if you look at Amazon, for example, and they're known of course for a certain management procedure, which is much more mechanistic. So if you look at management in general, actually Sophia.

It's never innovated since 1911 when the book on management systems was published, Principles of Scientific Management. It's still managing to not rock the boat, status quo, and over the years it has become, management roles have become data management roles. Picking boxes, making sure the metrics are fulfilled. That's obviously a job for an AI. Any AI can do this. So this is what I call the new MBA.

It's not a new degree, but it's management by algorithm. So management roles today, they've evolved into basically data management roles. They have to take boxes. They have to make sure the metrics are there. The metrics are being met. And this is what managers complain about. They don't have time to lead. They're basically managing and supervising data. And that's what an AI can do. I call this management by algorithm, the new MBA, not a degree, but what is really happening to the blue workers, as you said.

And Amazon is a great example of this. I mean, we hear there were several stories of people who felt like treated like robots. They were working on the assembly lines or they were supervising these teams, but because everything's so mechanistic and AI can easily take over. So management by algorithm is a reality today. And this emphasizes something. What is the difference between Amazon and then you and your job is what does it mean to be a human at work? The identity of a human at work.

work, you're a creator and that's, you can value a human identity because of his authenticity and creativity. Whereas in companies like Amazon and the examples here, we're providing here, how they work. We're reducing people almost to a data point. And that's where the balance is very sensitive and that you need to try to achieve as an AI seven leader. So yes, you need to know the AI, use it as a tool, but you need to reward people for being human still.

Sophia Matveeva (17:03.31)
because that's what people are looking for. They have fears. Well, they're facing uncertainty. Many companies are talking about the jobs of the future, but I don't see many companies actively investing in creating these jobs already. What you see happening mostly is that companies bring in AI and they fragment jobs because the job is seen as a number of tasks. Some tasks are being automated. And to your point, if it's a creative industry, that can work because then people create new jobs, new tasks for themselves because they can focus on it.

But if you see in more marketing, just more mechanistic management structures like Amazon, like Walmart, any company that fits that profile, you'll see that once the tasks are being taken away, it's very difficult in that context to create new dimensions in your job. If the organization doesn't invest in it. I wonder what investors really should take from this because.

Out of my audience, there are corporate leaders, there are startup founders, and also there are investors. And I think, you know, I'm going to blame the investors, partly for what's been happening, because I have seen this, I have seen equity analysts ask on earnings calls, ask the CEO what the AI strategy is and how they're going to use AI to, I don't know, cut costs and so on. And I think a lot of the time they...

the analysts themselves or the investors themselves don't really understand the question that they're asking. But then that puts pressure, especially on a public company CEO to then basically do something about it. So then, you know, every company now in their earnings report has to say that they have an AI strategy and how wonderful it is. And so I wonder, given this discussion, now that you have the ear of some investors,

What would you say to them, you know, when it comes to AI and what they should realistically expect of companies today? Well, you've said a lot of things that I agree with, first of all, but also very interesting, the notion. So let me start with the word that you used, AI strategy, huh? Companies need to report, they have an AI strategy and they need to report to their boards who even understand less than most business leaders.

Sophia Matveeva (19:25.422)
They really need to be educated because they're governing it, but they don't even understand. So AI is not a strategy. That's where it starts. And why am I saying this? Because if you say AI strategy, it means the primary focus is the technology. Humans come second because it's a new way of thinking and working. And that's what I describe in my book. That's what most people, business leaders think. It's a new way of thinking. It's a new direction. So what is a human centered perspective from that way?

we use AI, but humans are asked to think like an AI, like a computer. Whereas the truly human centered approach, but to achieve augmentation, what I said, enhance human intelligence, like in your job is no AI needs to be used, implemented and preferably developed in line with a human intuition to make sure they can excel. It's completely different point of view. That's where the education starts. Now, having clarified this, what business leaders should learn from this is that.

Why are they actually believing over hype? Because what are, to your point, tech startups, tech companies are scaring them because they want more investment, but they're scaring and they're making them happy at the same time saying, we have artificial general intelligence in three years. Of course people want to invest because they don't want to lose out. At the same time, the same person who's saying that will also be saying, yeah, but it may anti -humanity. So we also have to be careful here. So there's also certain fear.

So what is happening, who feels adequate in this world? No one except the tech companies. So we give them free reign and this leads to an AI arms race, for example. Now we see tech companies, basically we're being, we're witnessing tech companies trying to beat each other in every arms race they can identify and out of fear, but also out of hope, we keep giving them money. that's interesting. Now we have to bring it in. Now we have to bring it in, in the organization.

Yeah, but they know best. I mean, so what do I do? I put all my budget in, but they know best. We buy it. Now, what, that's one thing that you need to learn to see. The second thing is, again, this goes back to my earlier point. You spent just so much money because they're the only ones you can trust, but then you delegate to the tech experts. So there's no augmentation. There's only automation of tasks. And that on the long -term will undermine basically your organizational purpose. Again.

Sophia Matveeva (21:51.694)
Your biggest challenge as a business leader will be to align your AI comprehension with your organizational purpose. And the reason why I'm saying this is I've done several interviews with big companies and when I ask the top executives or board members, hey, you understand augmentation, don't you? And they say, yes, yes, yes. But then when I keep poking, they say, no, I really don't understand it. To be honest, we just have AI, it's cost effective. It can help humans. I said, okay, but what are you doing?

to make sure that AI can help humans. but that's not my job. No, I said exactly that's your job. But do you have the budget for this? no, we just bought everything. I mean, now the tech experts have to do it. That's why I hire talents and tech experts. Said you're forgetting one thing. The most expensive part of the AI adoption is the augmentation process, which means to train your people to learn to work with it.

create conditions that people don't have distrust and undermine the use of AI, reduce their fear, empower them, create flat communication cultures where they can give feedback because no AI model is perfect from its first run. I always compare it to a plane. If a plane arrives, it's full of data. You want to know what's in these data. And what happens then is they analyze this data. If there's something wrong with the plane, they need to make that decision. So what happens there is.

This information flows very quickly throughout the organization. Just imagine the normal business structure. It would take three months to get to the top of that information. A plane can't wait for three months. You would be bankrupt right away. But still, this is how we do it with any AI adoption process. So as a leader, you need to create these conditions. And that's the expensive part. Yes, you pay for the technology, but then when it's in the company, that's where you're going to invest. So two very big lessons.

So see what's happening and see what you need to do. So as my last question, I'm wondering what your advice is of kind of what is one practical thing that a business leader could do once they get to work. So there are lots of people who listen to this show on their commute while they're going to work. So literally you could give them some sort of insight or some sort of task that they could get on with the moment they get to the office.

Sophia Matveeva (24:15.534)
And what should that task be? So first of all, lifelong learning. You have to realize AI is, it's a tool in action, which means it's a tool. You have to learn how to use it, the way we explained it, but it's in action, keeps developing. So lifelong learning and it's easy. I mean, I know so many executives, just one hour a day. You read your LinkedIn, you take some articles or you register for a course, have that discipline, have that resilience.

Second thing is have a mindset of experimentation. Like I said, once you bring in AI, it's not going to be perfect right away. People have to be adjusted to it. People need to be able to give feedback. You have to train it, retrain it. And you'll find out that sometimes you won't even have to use it in certain settings. In other settings, you may need to use it. So train yourself into having that experimentation mindset. AI is not a perfect tool. Neither are your humans, but together they can create more.

So have that experimentation mindset. And thirdly, what is very important, humans first, AI second. No matter how much you automate, always give your people the feeling they're still being rewarded for being a human network. That is extremely crucial for the sustainable long -term success of your AI adoption process. So those are three simple things already. Thank you so much, David.

I've loved our conversation and thank you very much for being on the show. Thank you again for having me. I enjoyed it. Wasn't that interesting? If you found this episode useful, then leave the show a rating and a review. And if you're not yet subscribed to this podcast, then please do subscribe so you can carry on learning how to thrive in the digital age for free. And now that you've learned a bunch of useful stuff.

I wish you a wonderful day and I shall be back in your delightful smart ears next week. Ciao!