State of the Enterprise and the CIO, with Unisys CEO

Michael Krigsman: We're examining enterprise technology and transformation. Of course, that includes AI and implications for chief information officers. Our guest is Peter Altabef, the chair and CEO of Unisys.

Peter Altabef: Unisys today is about a $2 billion revenue company. About 80% of our business is integrating third-party software, putting it together, and then using it on behalf of clients. However, about 20% of our business is our own proprietary software that we license to clients and implement. So, we have a bit of a foot in both the solutions and pure software development worlds. So, we kind of do both.

Michael Krigsman: Now, Peter, we're all focused on AI. What kinds of investments in AI is Unisys making, and, what's the impact of AI on Unisys?

Peter Altabef: We track AI throughout the company. According to the most recent report I got, about five days ago, we have 135 AI projects going on. We do it at scale. We tend to look at AI through a couple of different lenses, Michael.

The first is, is it generative, or is it non-generative? Some people call it traditional. It's kind of weird to call it traditional when it's relatively new. Some people call it prescriptive, because it tends to be more exact than generative. But if we think about AI in history, Steven Spielberg came out with a movie called "AI: Artificial Intelligence" in 2001, so it's 23 years ago. And whether we talk about machine learning or deep learning, all of that has been around for 20 years. So when I use the term AI, I tend to combine traditional AI and generative AI. That said, generative AI, about three to four years old, has changed the game dramatically, so we can veer off into those.

When we think about AI in the company, we think about three lenses:

1. The lens where we're putting AI directly into our client solutions and where our clients will use the AI themselves. That, if you will, is the solution element.
2. The second element is where we're using AI in providing services to clients. I'll give you an example. When we think about code generation today, code generation today is making use not just of deep learning and machine learning, but generative AI. Coders are about 50% more productive than they were three years ago, because that's how much of a difference AI has made. It's our service to our client, as opposed to what our client does with our stuff.
3. The third element is almost back office. So it's how do you use AI and GenAI in the marketing department, how do you use it in finance, how do you use it in legal, and how do you use it in client development?

We tend to think of it through those three lenses.

Michael Krigsman: You mentioned that you have about 135 AI-related projects. What's the general state of maturity? In other words, to what extent are these projects proofs of concept or actually in production rolling out at scale?

Peter Altabef: We tend to measure them by proof of concept, pilots, or production. Those are the three categories. It's about a third, a third, a third at this point. So, it's moving remarkably quickly. And even GenAI is moving into production very, very quickly. That's true about our client solutions, working with clients. It's also true about how we're supporting our clients and even our back office. Pretty evenly spaced at this point.

When you're sitting down with a client, whether it's a CIO or a CEO or CFO, it's a bit unfair because you can drive value in many different ways. But it is accurate that you get their attention when you start talking about GenAI. All of the media attention has made them more aware. And so you'll get, if you think about how people are paying attention to you at a seven or an eight or an eleven when you get to GenAI, you get to eleven.

Michael Krigsman: What about the use cases and the levels of investment that your customers are making in AI today? What do you see with that?

Peter Altabef: We have put aside an investment kitty for anything a business team wants to invest in beyond their existing budgets. That investment kitty is available to people but requires a return on investment or an ROI. For the rest of the budget for AI, we have said, "Look, guys, get used to it being ordinary. Of course, it is like air; it's like water. Every company is doing this all the time."

So, while you can come to us for specific investments, rethink your existing investments because a lot of people have what I call the old John Lindsay approach, and I will date myself. John was the mayor of New York, and he was a popular mayor. And when he became mayor, he went to the police department and said, "We're going to cut your budget." And the police department said, "Well, we can't go on strike, but we will do whatever you tell us, but 100% of the budget cuts will come from policemen and women off the street." In other words, they would not cut their back office at all, which is the opposite of what he wanted.

And often, Michael, when you talk to a team and they ask you for more budget relief on a specific thing, it's always because it's the most valuable thing in their coffer, but they very rarely talk about the least valuable thing in their coffer. The real issue about budgets is you want to compare the new investment against the least important thing they're doing, not against the most important thing. That's the way we've approached AI. We really said to the extent you can, we really want you to fund this through your existing budgets. And that's a more disciplined approach than some. But we think that it's the right approach.

Michael Krigsman: That would be a slower approach as well because they've got to find the money. They have to prioritize the investment as they would any other project and stays away from AI being the shiny object that we must invest in today, no matter what.

Peter Altabef: It might be a slower approach if the CEO didn't get a report every two weeks about what every group was doing in AI. So we really have turned that on its head a little bit and said we expect you to do it within your existing budget other than for a special investment kitty. But by the way, your CEO is going to get a report every two weeks on what you're doing. So we think that provides a little more of an impetus to get it done.

Michael Krigsman: So, really, then the implication of this is a certain kind of investment discipline to make sure that you're not chasing wild objects without application to the appropriate business case. Would that be another way of saying it?

Peter Altabef: It would. When we started to test this and we started to report on it internally, I've got a report here. So in April of last year, we had 51 projects. As of last month, we have 135 projects. So clearly we're getting traction within the company.

Michael Krigsman: And what about your customers? Do you see the same kind of thing, that level of traction inside your customers?

Peter Altabef: Today we are. And when you did the introduction, you talked about CIOs, Michael, and I would say CIOs have historically been the most likely client for Unisys and still probably are, but to a much less extent because the extent to which you start to have solutions that dramatically move the revenue of the company forward, then all of a sudden your chief commercial officer cares. To the extent that you start having solutions that put your client at the forefront of their market, then all of a sudden the CEO cares.

And so it's more than about an incremental advance of technology. It's more about only saving money, all that part of the equation. You tend to move up into other elements of the C-suite, and that's important for us. And I think GenAI has probably been the poster child for a new technology that has been more seriously advanced by boards of directors than certainly than the cloud, certainly than if you go back in time, client-server. I mean, it's a technology movement that has had more higher-level interest than any I can recall.

Michael Krigsman: Would it be the case then that there is an understanding among the CEOs that you're talking with that AI and then GenAI specifically can lead to, can we say innovation breakthroughs or benefit as opposed to driving merely, quote-unquote, merely driving efficiency, meaning lower cost or faster speed, but still doing the same thing that we always have done?

Peter Altabef: What GenAI, in particular, has done even more than machine learning and deep learning is opened up the imagination. I mean, you really can think about your company and thinking about what you provide to your clients in ways that you could never even imagine before. And that imagination gets the excitement of C-level executives, it gets the excitement of many people inside companies.

And I think you quickly realize it's a different decision-making matrix because the imagination can only be fired if you understand the tool. So when we started this journey in February of last year, we started to say at our company, "Look, if you really want to highlight the art of the possible, you have to make sure everybody is educated in this." So we opened up our GenAI training not only to the people who currently had a specific business use, but to everybody in the company. And now between 96 and 97% of our company has taken GenAI training. Now we have more advanced training, but everybody is learning about GenAI already.

Then we had a bit of a mind-blowing idea for the more advanced training. Not only did we say that we created five personas, but each persona is based on what a job is in the company and how you can best use GenAI. We have five different advanced personas, but we opened up the training to everybody in the company. It doesn't matter whether your job is in one of those five, you can get trained in one of those five or all five because we thought this was a point in time where from a career tracking purpose, we want people to think about what the job they want, not the job they have. And that means getting trained in GenAI applications for the job they want to have.

So it's all about the imagination and it's all about the art of the possible. That, I mean, it differentiates this from some technology advances in the past which were useful but not as imaginative.

Michael Krigsman: As the CEO of Unisys, you're driving this awareness and training of GenAI deeply through the company. Why are you doing this? And how do you know that this is a good investment? Other than intuition? It seems like a good thing to do.

Peter Altabef: For a couple of reasons. First, as I said, when you talk to your clients, this is the subject that gets them most excited. They go from a seven to a nine to an eleven. Obviously, you want to have those discussions. Number one, the clients are most interested.

Secondly, the reality of it is this is where you can make breakthroughs. I mean, when we're talking with clients in travel and transportation or in retail or restaurants or banking, any of those sectors, the solutions you can talk to them about will move their business not only from a cost savings, but from a revenue generation. And if you want to talk to a client, you want to get them excited. You talk about revenue generation, you talk about new areas where they can expand. You can talk about them advancing over their competitors. Yes, they're interested in cost savings, they always will be. But this is a different dialogue. This is really a value add in a way that materially helps their business.

So it's not just the hype, Michael, it's the reality of it. The things we're doing for clients now using GenAI are honestly things that they could not do before. And it's opening up business advantages to them that they didn't have before.

Michael Krigsman: Directly addresses that innovation versus efficiency issue, that this is directly tied to innovation and therefore increase of revenue, customer satisfaction.

Peter Altabef: Whatever it might be, it really does. It's a good test for the viability of the technology. If the technology is only saving you money, that's okay. That's a good thing. And you can measure it on a pretty existing ROI scale. But if the technology allows you to expand your market share or expand your movement into different areas that wasn't accessible to you, that's a different kind of value. And of course, over time, first movers or second movers get to expand market share. And at some point that becomes defensive too, because the rest of the market catches up to you.

Michael Krigsman: Please subscribe to our newsletter, and subscribe to our YouTube channel. We have amazing shows coming up.

Let's take some questions from Twitter. And this is from Arsalan Khan, who's a regular listener. And Arsalan always asks great questions. And he says, from the top executive view, it might seem that companies are all in on, quote unquote, doing AI. But how do you become aware of resistance from the bottom up? Any lessons learned or frameworks?

And I was going to ask this question in a different way. But really the same thing is what does this kind of training do to the culture and the change that this is now driving?

Peter Altabef: I've already talked about two of the three ways we've tried to put this into our culture. One, obviously is you have our board as well as other boards and the C-level people asking about it. That's why I get a report. The second is you want to make sure that everybody understands it so that they can be imaginative. So that's why you train everybody in the company.

There's a third element to that. And the third element, I really give credit to Palo Alto. So we try to be a student of what other companies are doing, and we found Palo Alto really had a best practice in this, in the way that they kind of took a certain number of their middle-level leadership, if you will, and they really invested in that team. So they created very specific projects that were company-wide with high visibility and timelines and metrics. And they said, "We as a company are going to do this. And by the way, folks, we need you guys to lead this." And they have regular meetings of that, if you will, middle level of people that if you don't get their hearts and minds, then we call that permafrost.

So we borrowed that with their permission, and we have got that now in our company as well. And I will tell you, that's been a little bit of the secret sauce between a C-level effort and a homegrown effort. You have to get that permafrost to be a force of good as opposed to be a force of resistance.

Michael Krigsman: How do you accomplish that? Resistance to change is an endemic part of any transformation effort, regardless of what it is. After all, it comes back to human nature. So how do you do that?

Peter Altabef: The people leading those company-wide efforts are not C-level people. The people leading those company-wide efforts are, if you will, those middle folks who get an opportunity to get very, very high profile. And it's not one person. We have nine of those. And for each of those nine, we've created a group of people from around the company that are actively working together. 

So I hate to be, I don't want to put a false note on this, but you have to get them enlisted and you have to make them part of the solution. As I said, every company will do it differently, but that has really worked for us and we think it's a really good effort.

Michael Krigsman: If you don't get the body of folks inside the company behind you, then ultimately, any type of major transformation effort is not going to succeed. And you may not even know why exactly. You just know things are not working. Why isn't this happening the way it should? Because it makes so much sense.

Peter Altabef: Keep in mind, a company like ours, Michael, not only has to do it for our own efficiency, but if we don't do it effectively for our clients, it affects our ability to expand our revenues, it affects our ability to expand our margin. So one of our clear tests is, are we bringing those solutions to our clients, and are our clients really embracing those solutions?

I will say this is also new territory, because usually for companies like ours, you go to a client with 80% of the solution figured out, and then you customize it for the client. That percentage is probably a little less here, because not only is the solution have to be customized, the imagination has to be customized. You have to understand your client could be thinking about something you haven't even realized, but you know how to use the technology in a way the client may not. So it really becomes, you know, I mean, what we like to say is that there's no "not invented here" at our company. And I think that that's true for most good companies. And it's a very roll up your sleeves and work at creating something new with your clients.

You have an efficiency standpoint. If everything is brand new every time you walk into a client, it's not very efficient for you or the client. The client expects you to come with a methodology, with IP, with a solution. But more of the solution is now custom-made because of this imagination quotient than we've seen in the past.

Michael Krigsman: This is from Chris Peterson on Twitter. And Chris is another regular listener who also asks great questions. And Chris says, the journey from 50-ish AI projects to 130-ish in 18 months is interesting, but how many have been completed and achieved their ROI projections in that time?

Peter Altabef: We have about a third of those in production now. And to get into production, you have to hit those ROI targets. So I would say that the track record is pretty good, and the 135 are the ones that are active. So there have been projects in that journey that fell away that are no longer counted in the 135 because they're not active anymore.

One of the things I would say to that question is, I know the current hype is about AI, and I know the reality is that AI can deliver on its hype, but there is always something around the corner. And I think it's really important not to get completely enveloped in the current, because especially for technology companies like ours, around the corner matters a lot. So when you're thinking about training people and you're thinking about solution development, we tend to not only look at what's in front of us, which is GenAI, but also what's around the corner.

Michael Krigsman: You're a public company, and so every quarter you're reporting to your shareholders and to analysts who are ultimately interested in financial performance and less interested in the AI-centric view that we're talking through today, which I suppose creates another layer of challenge as you're thinking how to balance all of this.

Peter Altabef: They are and they're not. There's an old adage that if you follow analyst desires, you will follow them right off a cliff because you will always be doing the things that increase your revenue or your profit this week, but it means that you won't be investing for the future. So you really have to do a combination of both. And the best analysts know that really, really well. So the best analysts are not only talking to you. And that's true for investment analysts, it's also true for industry analysts, whether it's a Gartner or Everest or any of those really quality analysts.

So we tend to have to mix in. The best company is not only looking at what's the art of the present, but also what is around the corner. For us, the art of the possible around the corner is quantum. And we're doing quite a lot of work in quantum now. Even though if you talk to most CEOs, they assume that that is four to five to six years away. In some respects, it is. And in some respects, it's available right now. And most people don't fully understand that.

Michael Krigsman: On the subject of quantum, when you say it's available right now, can you describe some practical use cases? Because we all, as you said, we all hear in the background, quantum is coming and it's going to change everything. But we don't hear about practical use cases very often.

Peter Altabef: NVIDIA is, of course, a company that a lot of people are paying attention to and looking at their market cap. But NVIDIA really, that did not happen overnight. That happened because they had the foresight to really decide that they were going to create chips that were extremely available for generative AI use. And they didn't create those chips overnight, nor have people been able to replicate what they did overnight. So it really is a question of, it seems like a little bit of an overnight success from a market valuation standpoint, but it's anything but an overnight success. And so that is why companies, I think, really do have to look around the corner.

So, on the subject of quantum, I kind of think of it in three buckets:

1. There is what most people think of in quantum as general-purpose quantum computing. And when you read articles, most quantum is about that. So you hear about qubits, and there are multiple countries as well as companies that are getting more and more advanced stability with more qubits. That's the way you kind of measure a general-purpose quantum computing. The most general wisdom is that we haven't, as a society, get to the level of stability we need for really functional general-purpose computing that is quantum. And that question is, well, when does that happen? The standard answer used to be 15 years or 10 years. The standard answer now is, we really don't know. This is an indication that it may be sooner, but we don't know.

   General-purpose computing is an interesting idea, and Unisys has played a role in it. In 1946, the first general-purpose digital computer was the ENIAC, which was built by Unisys. The first commercial computers were the UNIVAC and a similar competitor from IBM in 1950. So we're headed into another world where you will have that general-purpose quantum computer, but it's not here yet.

2. The other two elements are more current. And so, let's talk about another quantum computer called quantum annealing. So, quantum annealing is a subset of quantum that is very specific for optimization problems. Here, you have already achieved sustainable qubit numbers because it's not everything to everyone. It's very, very specific. We at Unisys have taken quantum annealing, which is already mature, and we have used it in optimization solutions, our logistics solution, which is called Unisys Logistics Optimization. What a name! Really takes advantage of quantum annealing. Optimization is something that can be done right now with quantum computing. So that is the second of the three categories.
3. The third category is what we call post-quantum cryptography. And post-quantum cryptography is defensive. Quantum, that is, using software and using different algorithms to protect your data or protect your software from the coming onslaught of quantum computing. Because the general perception is that when quantum computing becomes available, current levels of encryption in most encryption will be at risk. So, you want to protect yourself for when quantum becomes available.

And the key question there is, why bother now? So why do that today when quantum may not come for some number of years? The answer is that bad guys are stealing your data today, and bad guys are stealing your software today. And although current encryption may be sufficient against classical computers, it won't be sufficient against quantum. So if something gets stolen today, will it still be relevant when they can take that and break it in the future? So the current version, the current view is you want to protect yourself as quickly as you can so that the things that are stolen today or next year are not available to the people who want to break that encryption when they get quantum computing.

That's the third element of what people are talking about in quantum. NIST, which is putting forth standards in post-quantum cryptography for the future, has actually been working on this for about four and a half years, and they have said that they intend to release standards this summer. So this is quite current.

Michael Krigsman: So, for organizations that do want to start preparing right now, what should they do?

Peter Altabef: That first level of general-purpose quantum is still a bit beyond reach. The second level of using specialized quantum, such as quantum annealing, is available right now, and we're taking advantage of that. So if you have an optimization issue or challenge, whether it's in logistics or financial services or even retail or healthcare, you can solve optimization questions using quantum now.

And then post-quantum cryptography is something that some companies already put in place. So Apple is putting it in place, Microsoft is putting it in place, but we're all going to have to do that. And some companies have started doing that ahead of the NIST standards. Others are waiting for the NIST standards to come out. But those standards, we're close to the end of the summer, so those standards really are expected to come out very soon.

Michael Krigsman: Let's talk about digital transformation, which I realize is no longer the sexy topic it once was. Now it's been replaced by AI, of course. However, there is an enormous amount of money that is being invested in transformation. You just described one form of transformation, which is logistics. But there's also transformation in the workplace. Do you want to talk about that a little bit?

Peter Altabef: It may be the biggest challenge that I have as CEO of a company, and I think probably the biggest challenge of most companies. The challenge there, Michael, is fear. To be very simple, even technology companies have to deal with people who are fearful of their job becoming redundant. They don't know what's going to happen over the next five to ten years. If you're about to end your career, those people tend to be less fearful. If you're just starting in the job workforce, you tend to be less fearful because you're current and feel like you can catch the wave. But if you're somebody who's been in the market for 20 years or 25 years, you're pretty fearful in general.

And so, I think one of the jobs of a company, whether it's ours or anybody else, is to lower the fear. And you do that by educating, you do that by training. You do that by emphasizing career mobility inside the company. You say, "Okay, well, these jobs are opening up, and how are we going to get you trained to take advantage of that new job?"

Interestingly, AI is critical for that effort, too. So, as you think about workforce management, you think about how you get the data and how you process the data to know what jobs are coming online. How do you train people for those jobs, and how do you match? AI is perfect for that. So interestingly, it's using generative AI to help you lower the fear for people who fear what happens to AI in their jobs. So it all makes sense, but it's not initially intuitive.

Michael Krigsman: We have another question from Arsalan Khan on this topic, and he says organizational transformation requires a balance between strategy, politics, innovation, culture, and execution. SPICE. Since you are at the top, how do you make people aware that politics can derail potentially successful projects? And let me broaden that to just the general question of politics. And I think it also goes back to forces against innovation that we were talking about earlier.

Peter Altabef: If politics means people in the company who are seen as political in the sense that they're seen as promoting themselves and using their influence to get other people to say good things about them, a good organization, in my opinion, has no tolerance for that. You can't allow people to use the organization for their personal gains. It really has to be everybody rowing in the same direction for the company's benefit. And you kind of know when you're at a company, in my experience, whether the company allows those kind of people or has no tolerance for those, and we don't have a tolerance for those. There'll always be some. And you basically say, "Sorry, that doesn't work here."

The broader question about all of those different elements which are completely accurate, and how do you actually manage all of those elements at once? My advice would be, don't get too cute about it. Be as transparent as possible because you can't manage all of those variables at the same time. You have to, but you can't possibly be successful if you're not being transparent. So this is an old adage of Winston Churchill's of tell people what you're going to tell them, tell them, and then tell them that you've told them, and that sounds repetitive, but it's really not. 

And I think as long as you're transparent with the company and the people in the company is where you want to go, where you want to take it, and everybody says, "Okay, I hear you. And so I'm not going to be surprised." I think people will come along with that. I really do.

Michael Krigsman: As CEO, the management or the encouragement of culture ultimately falls to you. What do you think about that aspect of your role?

Peter Altabef: Different companies are organized differently. I believe in a servant leadership culture. I do believe that the more people you have, that you have direct responsibility for, as well as the technically higher up in the organization, the more you have to spend your time helping the organization, helping people succeed.

We just announced earlier this week that two people in our organization are getting new roles that are significant advancements for them. So I sent a note to those two people saying, "Okay, I'm here to help. How can I help you?" And that's the first instinct I have: How do we help the people in the organization succeed? That's an element of servant leadership.

When COVID occurred in 2020, I made a personal commitment that I would have a video call one-on-one with everybody in the company who came down with COVID. And I had, I don't know, 2500 calls. In hindsight, that took a lot of time. But it also, I hope, showed the organization we deeply cared about how this affected every individual. So I don't know if that was the right thing or the wrong thing to do. But I do think that how you show servant leadership is really important for a company's culture. We call that our "let me help" culture here, which is about helping each other.

[Editor's note: Following the live interview, Peter Altabef sent us the following clarification, "We had about 2,500 people who contracted COVID-19, and while I tried to talk with as many as possible in one-on-ones, I didn’t make 100%. After more than a year, I dramatically slowed down the effort. While our HR group could give me insight into the individual’s journey with COVID, I was ill-prepared to handle situations of family members who had passed from the disease."]

Michael Krigsman: So that "let me help" culture is something that is defined. People are aware of it. It's accepted inside the organization, and people expect that's the way it is.

Peter Altabef: We live in a world of instantaneous communication. We live in a world of town halls where the entire company can be in a town hall. We do those every quarter. But, you know, every once in a while, you have to sit back and say, how do we do it in an old-fashioned way that is more powerful, and people pay attention.

I have for years and years now, had a monthly email, which is my way of communicating to the company every month. And it's what's on my mind. Sometimes it's about things outside the company; sometimes, it's about the company. But it is a one-to-many communication and people respond to that email, and I told them, "Hey, I'm going to get back to you within 24 hours" because it has to be not a send communication, but a dialogue. But it's very personal. And I think the more we have impersonal communication, the more important it is to say we can rise above that and keep it very personal.

People stay at a company because of the relationships they make. We had a celebration earlier this morning of someone who was retiring from the company after 42 years. And I got on the call and was on it for an hour. We had people throughout the company who had worked with this individual in his 42-year career. That was important for the person, but it was really important for the other people on the call to see how much we celebrate that kind of thing. So I think making it personal matters in a fairly impersonal world.

Michael Krigsman: So, I'm assuming it would be fair to say that the maintenance of culture, the demonstration of servant leadership, and also leading by example is very important to you, and therefore you're willing to invest a significant amount of your personal time in that.

Peter Altabef: A series of books are written by an author named Pat Lencioni. And I like Pat, and I've kind of become a little bit of a student of his because he has this idea of a team one. And I've used it at all the companies I've been privileged to be associated with. And it's an interesting element. So you take my direct reports. So the CFO, the head marketing officer, the head of HR, the head of legal, and the head of technology, the head of operations, and you take that team and you ask them who their team members are. And most of the time, Michael, in most companies, even in ours, in an off-guard moment, they will say, "Well, I'm the head of marketing. I have x number of people. I have y budget. That's my team." That's an answer, but it's not the best answer.

The best answer is for that head of marketing to know that his or her team are the other people who report to the CEO. And if they think and act like the other people, whether it's the CEO or the CFO, if that senior team thinks that it is their team one, the whole company will get that message. And all of a sudden, what used to be stovepipes will be very, very flexible, because all of those other organizations will be very, very comfortable moving laterally into these other organizations and communicating. And what you really want from a company is not stovepipes. You really want that lateral communication and that does start at every level. So when I go to my chief financial officer and I say, "Who are your best team members?" And they refer to the head of operations, head of marketing, and head of finance and head of HR, that's the answer I'm looking for.

Michael Krigsman: You want that organizational fluidity which will then automatically if it's done well, break down the silos.

Peter Altabef: That's right. Ultimately, you want an organization that not only can react quickly but can leverage itself so that you, on a project-by-project basis, may have a very senior person in the company on a project working with a much more junior person, and that junior person is a project leader. So, inside our organization, we don't have titles. We don't talk about vice presidents or senior vice presidents. We do that in press releases to the outside world. But inside the organization, you won't hear us talk about titles. You'll hear us talk about function and what a person functionally does. But we try to avoid titles if at all possible, which is a...

Michael Krigsman: Step towards helping create the psychology and the culture of flattening that organization.

Peter Altabef: Correct. Because it looks odd for a senior vice president to report to a director on a project. Well, so they're not. Don't call them that. Just have them as people, and suddenly that becomes a much...

Michael Krigsman: More effective project, doing whatever makes sense.

Peter Altabef: For that project and having the leader as the right leader, even if the leader is more junior, which it often is.

Michael Krigsman: This is a question from Lisbeth Shaw. And Lisbeth says, how important is it for an organization to simplify its enterprise technology? How can they do so with minimal disruption and risk while continuing to operate, especially when running a mix of legacy and modern technology?

Peter Altabef: Legacy technology, we tend to call now technical debt. That's the term we tend to use for that. Our chief information officer inside our company deals with that every day. Where we have come to consolidate. We have a consolidation project, and we reported to the board about it last week, is really; how do we work through that to get the objective of simplification correct? It makes cybersecurity easier to protect. It makes efficiency easier to monitor. It makes rolling out new generations of technology easier. It's just better.

What we tend to do at the company is give people a warning. We say, "Hey, we have ten software programs that do about the same thing, but in three months, there are going to be two, and you can pick from the two, but you have three months to move from one of the other eight over into the two and that's all you get." We find that as long as you give people some reasonable notice, they have some opportunity to move into the programs that are really your go-forward programs. And that's the way we tend to work it. But you have to be very thoughtful about it. And if you don't have a very specific plan to do that, you will be stuck with technical debt for a long, long time.

Michael Krigsman: We have a question on LinkedIn from Greg Walters, who says, what is your perspective on the current AI disillusionment versus the AI hype? And is this disillusion itself hype? So, AI disillusionment versus hype? And I'll ask you to answer relatively quickly because we're just going to run out of time.

Peter Altabef: The answer is yes. So AI hype had part reality, part hype. AI disillusionment is part hype and part reality. There is no question that in the long term, AI, including generative AI, will significantly change the way a lot of companies work and the solutions we give to consumers and other companies, it is going to matter.

I do think going back to my "around the corner," one of the things that will take AI hype and really put it on steroids, if you will allow the analogy, is the combination of AI in its current form on a computing platform that is much more powerful. You've already heard me talk about AI combined with stuff in the future will actually deliver more than we have today. But even AI on classical computing is real and ignore it at your peril.

Michael Krigsman: Let's jump to CIOs. As CEO of Unisys, I know you spend time with CIOs, with CEOs, CFOs. In this environment of AI driven by AI, where does that leave the CIO? What is the CIO role today?

Peter Altabef: The CIOs now have stuff in their domain that is of great interest to the CEOs and CMOs and the CCOs and the CFOs. So this is the time for CIOs to shine. This is the time not to be defensive. This is the time to show your stuff. And I will tell you, this is a time. One of the great skills of a superior CIO is not to get trapped in engineering speak and not to get trapped in functions, but actually to communicate in a way that a non-expert, the rest of the C-levels, can understand the value and opportunity you are providing.

So a CIO is almost a translator from all of the specific engineering skills that they have. CEOs don't want to hear about technical debt, but CEOs do want to understand what are the business solutions that advance the company if you can be successful in a good consolidation project. This is a real moment for CIOs.

Michael Krigsman: It's difficult for a CIO to fully put in place and think through the technology strategy that will support revenue or strategic goals of the company. Because the CIO's time is spent on the tech, not the business strategy.

Peter Altabef: What we have done with our CIO and I would suggest most companies take 25% of that person's time and get them out of dodge. So have your CIO talk to your clients, have your CIO talk to the rest of the industry.

Don't be so introverted because I think the skills you get from talking to clients and meeting with clients and prospects are the skills you need as a CIO to talk to your CEO and to your board. So I agree with you, Michael, but the answer is to get out of the office and start spending more of your time in the outside world.

Michael Krigsman: So, for a CIO, however you do it, you need to have that broader lens.

Peter Altabef: And that's the difference between a C-level executive and a non-C-level executive. And it's time for the CIOs to take their place at the C-level. It really is.

Michael Krigsman: We have yet another question from Arsalan Khan, who says he read somewhere that the average lifespan of executives is about three years. How do you make an organization aware that it's about the idea and the strategy, even after the executive has left? And we've all seen great projects that died on the vine because the sponsor left, and all that investment was wasted.

Peter Altabef: I don't have an answer to that. I can tell you that the data is right. The average lifespan is three to five years. That is not necessarily the best answer for a company. So, I mean, I talked about an executive we had who left after 42 years this morning. It wasn't a C-level executive, but somebody who made a difference at the shop. And so I would tell you, I kind of rue that. And I don't have an answer for that other than it shouldn't be the right answer. I think if you have a culture where people are actively engaged and they see the contribution they're making and they're able to make that contribution, you should do better than three years.

Michael Krigsman: What advice do you have to folks working in an organization who could be middle managers on being successful and developing their career from your point of view as CEO?

Peter Altabef: Be excited about what you do and be willing to change what you do inside an organization. To be excited. Most people hesitate because they're good at something and afraid of not being great at the next thing. So, they never evolve. Whether you're changing companies or working inside a company, evolving yourself and your skills is critical.

So, you know you have to evolve. If you move to another company, well, evolve within the company. And if you're in the company and you can't evolve, and you wake up in the morning, and you're not happy with what you're doing, go somewhere else. Find a place where you can be happy because life is too short for you not to be happy, but give your organization a chance because they most likely want you to stay. And you could probably have a better career inside that organization if you're willing to evolve. But if it doesn't work for you, then move on.

Michael Krigsman: I love that. It's great advice to be open to change and be willing to change and adapt.

Peter Altabef: Yep. Well, thank you, Michael.

Michael Krigsman: With that, we are out of time. A huge thank you to Peter Altabef. He is the chair and CEO of Unisys. Peter, I'm very grateful to you. Thank you for being with us today.

Peter Altabef: Thanks for having me.

Michael Krigsman: And a huge thank you to everybody who listened, especially those who asked such great questions. Before you go, please subscribe to our newsletter and subscribe to our YouTube channel. We have amazing shows coming up.

Check it out. Check out cxotalk.com. Thank you so much, everybody. Have a great day. Bye.