Q&A: Dell CEO Michael Dell on the Future of AI - McKinsey

Michael Dell, CEO of Dell Computers, is the 15th richest person in the world with a net worth of US$80 billion. Not bad for someone who founded his company while still at college in 1984. 

Dell was born in 1965 in Houston, Texas. He began his entrepreneurial journey at the University of Texas, where he started selling computer upgrades from his dorm room. Dell dropped out of college to focus on his burgeoning business, which he incorporated as Dell Computer Corporation that same year.

Under Dell's leadership, the company grew rapidly becoming a major player in the personal computer market. However, Dell faced challenges in the early 2000s as the PC market matured and competition intensified. He stepped down as Chief Executive Officer in 2004 but returned to the role in 2007 to lead a turnaround.

Dell took the company private again in 2013 in a US$25 billion deal and then led the acquisition of EMC Corporation in 2016 for US$67 billion, creating Dell Technologies.

Dell Technologies has evolved into an end-to-end provider of artificial intelligence solutions to enterprises, 72% of which have now adopted AI compared with 55% in 2023.

Here, Michael Dell talks to McKinsey senior partner Tarek Elmasry about his belief that AI is a trend like no other, one which will transform the way organisations operate, organise themselves and do business.   

McKinsey: You recently said that we are merely in the pregame for gen AI. As we make the leap from computation [computers that calculate] to cognition [computers that think], how far do you think we can go over the next five to ten years?

Michael Dell: About 60 years ago, we started this journey of calculating computers, and that’s basically what we’ve been doing ever since. Now, with the breakthroughs in AI and gen AI, we’re moving into cognition and intelligence and unlocking the power of data with all this innovation. It’s the latest point in the curve on Schumpeter’s creative destruction, but it’s accelerating it enormously. The ability to reason using all the knowledge in the world on any given subject and more accurately distil that into outcomes and new understandings will accelerate and advance scientific discovery. It will make humans happier, healthier and more successful. It will improve everything we humans try to accomplish. I’m incredibly excited about it. As with any new thing, there are all sorts of uncertainties and questions, including how’s it all going to happen. Nobody knows, and we love being in the middle of it.

McKinsey: What exciting things are Dell’s customers doing with AI?

Michael Dell: The fun thing about your question is that almost anything interesting and exciting that you want to do in the world revolves around data. If you want to make an autonomous vehicle or advance drug discovery with mRNA vaccines or you want to create a new kind of company in the financial sector, everything interesting in the world revolves around data. All of the unsolved problems of the world require more compute power and more data, and this is why I love what we do.

This morning, I was meeting with a pharmaceutical company. We were talking about how AI could help them in drug discovery, in their process of finding people for drug trials in their documentation process, in their supply chain forecasting, and more. Like most companies, they’re just at the very beginning. We weren’t talking about this two years ago. AI and machine learning and propensity models and the like have been around for a while. But everything is getting a refreshed look because of all the attention that generative AI has received.

McKinsey: Organisations have more and more of that data, and it all needs to be managed. Almost all of them believe that their data will be hosted both in the cloud and on premises. Nevertheless, there’s a debate about whether one alternative is better than the other. How do you see this playing out?

Michael Dell: We believe that most data will be created in the real, physical world, and that most actions taken on that data will happen in the physical world as well.

A data centre or a cloud is a place where you take that data out of the physical world to do something that makes it more valuable. That’s not going to go away. But you might also create value from your data in a place that you’re renting, like a colocation facility. And you might do it in your own facility. There are many, many ways of doing it.

Our job is to connect all the pieces together, giving customers the best options and choices. Organisations zealously protect their most valuable data — it might be their most important intellectual property. It is an incredibly valuable asset. So companies don’t just send that data off to a provider and say, “Sure, you can have all of it. Do whatever you want with it. Sell to our competitors, train your models, and we’ll just send you all our money till we don’t have any more.” People have figured out that’s not a great idea.

The next frontier for these kinds of technologies is unleashing the power of that data so that organisations can be more productive and efficient, and even reimagine themselves. Now we’re moving toward that kind of capability. We’re seeing all sorts of processes that could dramatically improve outcomes, whether in software development or customer service or sales or documentation or marketing, and so on.

McKinsey: How do you think gen AI should be regulated without constraining innovation?

Michael Dell: I empathise with regulators. It’s an incredibly hard job when technology is changing so fast. You could even say that any time you put a regulation in place, chances are it might look kind of silly in 12 to 18 months. People will ask, “How could you not know this was coming?” But of course, you couldn’t have known what was going to happen. There need to be some guardrails and regs, but in the marketplace of countries, you’ll see a range of approaches from let it rip to it’s a bad idea, don’t do it. When Edison came out with the light bulb, the candlemakers said, “This is horrible.” The defenders of status quo never like change and progress.

There’s always a fear that these technologies are going to replace humans. I think that’s quite wrong. AI is going to augment humans, not replace them. If I’m a knowledge worker in an organisation, I want the best tool possible to help me do my job. If I have a task or a meeting or anything else I need to prepare for, I want a tool that helps me find all the pieces of information across my entire digital platform that are relevant. AI that augments people gives companies many more options, and a lot of flexibility. It makes the world better, increases opportunities, and expands economies. I think AI will lead to a fantastic period of global economic expansion.

That said, we do need protections. But these are questions that society has to answer. The technology companies can’t answer these questions. They are the ones and zeros. Maybe that’s a cop-out, but I don’t think so.

Here’s a way to think about it. A computer comes out of our factory and is going to a person. It could be a big computer, it could be a small one. What is that person going to do with this tool? Maybe they’re going to cure cancer. Or maybe they’re going to do something bad. What that person could do on a digital scale is scary. But we don’t know. The defining element is the person. We have a lot of laws and regulations that prevent people from doing bad things. Those same laws and regulations should be applied here.

McKinsey: Some people are starting to wonder whether gen AI is being overhyped. What’s your take?

Michael Dell: Any breakthrough thing of significance, whether it’s technology or not, is by definition going to be overhyped. There will be outrageous claims and ridiculous projections, and a wide range of outcomes. But directionally, everything points to something pretty big here. We know it; we see it with our customers. We see it internally. But yes, absolutely, there are going to be spectacular failures along with the spectacular successes. We’ve seen this movie before.

McKinsey: Over your 40 years in the business, you’ve seen value shift repeatedly between various parts of the technology industry. How do you see value shifting over the next five or ten years?

Michael Dell: You just can’t predict that. If you were to go back to any period of time over the last few decades, you’d see that no one knows.

What we focus on is constantly getting better at the things we’re really good at. If we solve important problems that customers have now and in the future, we will be successful, and customers will reward us with new business. If we don’t, we’ll go away. We won’t be here anymore. If organisations give their people the tools they need to succeed, employees feel much better about that company. And the opposite is also true. Nobody wants to work at a Dilbert cartoon and work with junk. They’re smart people. They know what tools are available.

McKinsey: It’s been said that in race car driving and swimming, you win during the turns. You don’t win on the straightaways. If you look back over the history of Dell, what are the big turns that you have navigated well?

Michael Dell: Well, I’d say we got a lot of them right, but we didn’t get all of them right.

McKinsey: Well, you’ve been at the middle of two or three of the most profitable tech deals ever, including the decision to take Dell private and the US$67bn acquisition of EMC and VMware, in 2016. At the time, that was the largest tech deal ever.

Michael Dell: There’s an interesting side note to those. Over the past 12 years, there were three times when the ten-year Treasury note was below 2 per cent. The first time was when we went private. The second time was when we bought EMC and VMware. The third time was during the coronavirus. When money goes on sale, stuff happens.

One of our major turns was certainly the launch of microprocessor-based servers. We started with an original idea: selling PCs direct to the customer. That turned out to be a pretty good idea. As microprocessors got bigger and stronger, you could build client servers. So we went after that in the mid-Nineties.

Then the internet moved to the World Wide Web, which was a dream for a company that sells directly to the customer. Cloud computing and a more software-defined infrastructure led to incredible growth in data and the importance of data storage. Then buying patterns shifted from customers buying the ingredients of a network of storage servers to customers wanting a whole system. Customers wanted services from us and more capability. We managed all those turns. Mobile phones, on the other hand, were a turn we didn’t figure out — even though we saw it coming.

McKinsey: What do you take away from the mobile phone experience?

Michael Dell: We are not an operating system company. It doesn’t mean we won’t be 50 years from now, but right now, we’re not an operating system company. It’s not what we are good at, and the defining success factor for the mobile phone was the operating system. Android and iOS won and there wasn’t really a way to make a decent return on investment by making Android phones. [Dell introduced its Venue and Venue Pro smartphones in 2010 but discontinued production in 2012.] So that turned out to be a bad use of capital.

But the mobile revolution also accelerated the growth in data. When you use your mobile phone, you create data. To handle all that data you’re creating, you need servers.

McKinsey: People have talked about the demise of the PC. Now they’re talking about a supercycle driven by a shift to generative-AI-powered personal computers. How could the PC change over the next three to five years?

Michael Dell: When we talk about the PC, we’re really talking about a device with a large screen and an interface that you can easily use to analyse and manipulate large amounts of information and data. And now most of the AI that you need will run on your PC or phone.

Many of these devices will connect with some bigger, more complicated model [via the cloud], for security reasons, say, or because you need more compute than is on the device itself. Again, knowledge workers want the best tool possible to help them do their jobs. If I’ve got a PC with not just a CPU and a GPU [graphics processing unit] but now also an NPU [neural processing unit], it can help me in new ways. This AI-powered PC can look at all the relevant pieces of information across my entire digital platform that I need to understand to be ready for this meeting. The ultimate killer app of the AI PC is going to be saving you time and making you more productive.

McKinsey: In your books, you talk about the importance of relationships, customer centricity, integrity, and maintaining an entrepreneurial spirit in the face of adversity. When you look back over your time at Dell, what are the most surprising or best leadership lessons you’ve learned?

Michael Dell: Yes, I would say it’s all those [that you just mentioned]. Those are so powerful for us. They’re built in — we do those things based on experience and pattern recognition and knowing what works and what doesn’t.

Speed to change is another key to success. In the tech industry, we often say change or die. There’s the quick or the dead. But I think it’s increasingly true across a broad set of sectors.

And it’s one of the reasons I’m optimistic, because we have so many interesting opportunities to refactor things. Our company is built for changes like that.

McKinsey: You have employees operating all around the world. Your product mix is increasingly diverse. Customers’ needs are getting more complicated because they’re not just asking you, ‘Does the box do what I think it does?’ Now they’re asking, ‘Can you help me do all the things I want that box to do?’ Given all that complexity, how do you think about increasing the speed of decision making at Dell? Is that something you worry about?

Michael Dell: We do worry about it. There are a couple things here. One is that we love the customers who want to go really fast because they’re challenging us to go at speeds we’ve never gone before.

We also want to be very intentional about identifying which activities are the most important for the company, and quadrupling down on those. We want to be the best in the world at those. And we’d like to do less of the things that don’t rise to that level. That way we can give time back to our people so that they can more quickly do more of the things that matter.

How do you have an organisational construct that allows you to go at very, very fast speeds? Part of that is having a very clear strategy and not having ambiguity about what the priorities are. But we also need to have defined units of responsibility and accountability, and not have all the decisions come to the centre. You can’t manage a big organisation that way if you want to go fast.

McKinsey: You’ve had a really special set of partnerships with people like [Dell COO] Jeff Clarke. What have you learned about yourself through the process of working hand in hand with such critical managerial allies?

Michael Dell: I know I can’t do much of anything by myself, so I’ve been intentional in trying to find the best people I can to help me. And fortunately, I’ve found some great ones. I think it’s super important. You can’t really do much by yourself. Why would you try? That approach never made a lot of sense to me. And yes, it’s kind of the multiplier effect. If you think about all the things that you want to accomplish, you’ll need a lot of talented, energised people who bring diverse and interesting perspectives to what you do.

McKinsey: What advice would you give a 14-year-old today?

Michael Dell: I’d say make sure you study math and science. I’d ask them what they’re interested in. Hopefully they’re passionate about something, and hopefully that something’s going to help other people.

At 14, the parents matter a lot. If you’re the parent of a curious child, let them run with that curiosity. Feed it. Don’t quash it. My mother had a cedar closet in the house. Once, when my parents went out of town, my brother decided that it’d be a good idea to take all my mother’s clothes out of the cedar closet. I helped him a little bit, but it was mostly him. We took all the clothes out of the cedar closet and turned it into a darkroom to develop photography. Now, a lot of mothers would have been pretty upset. And while my mother said, ‘Well, it wasn’t such a great thing that you took all the clothes out of the cedar closet,’ she recognised that her son had done something amazing at a young age and didn’t quash the curiosity.

I was always taking apart the televisions and the telephones in the house, anything with electrons running through it, because I wanted to know how they worked. My parents could have gotten really upset with me, and sometimes they did. But they let me run with my curiosity. The parents play a pretty big role

McKinsey: Now it’s time for the rapid-fire part of the interview. Just say the first word that comes to your mind. Barbecue or Tex-Mex?

Michael Dell: Barbecue.

McKinsey: Read a book or watch a movie?

Michael Dell: Read a book.

McKinsey: Football or baseball?

Michael Dell: Basketball.

McKinsey: Mountain biking or cycling?

Michael Dell: Cycling.

McKinsey: If you weren’t an Austinite, where would you live?

Michael Dell: Somewhere in Texas.

McKinsey: If you weren’t an entrepreneur and CEO, what would your backup profession be?

Michael Dell: Gee, I don’t know. Maybe unemployed.

Michael Dell was being interviewed by Tarek Elmasry, senior partner in McKinsey’s Bay Area office

This interview will appear in the October issue of McKinsey Quarterly

******

Make sure you check out the latest edition of Business Chief, the monthly digital magazine for CEOs of multinational companies

******

Business Chief is a BizClik brand