DXC’s Chris Drumgoole on Infinia ML’s Machine Meets World

DXC CIO Chris Drumgoole on Why Human Oversight Accelerates AI

Join Infinia ML’s Ongoing Conversation about AI

James Kotecki
Nov 17 · 9 min read

Episode Highlights from Machine Meets World

This week’s guest is Chris Drumgoole, CIO at DXC Technology.

“Even though the raw technologist will feel like the use of human oversight’s going to hold them back, in the reality it’s going to make the technology adoption go faster because it’s going to get people more comfortable with it. They’re going to understand it better. And then they’re going to start to trust it.”

“In the modern connected world, a bad piece of AI could make a really bad impact, really wide and really fast. So I think a little more human oversight while it learns is the way to go on that front.”

“For the last God-knows-how-many years, everybody was talking about Big Data…and the world is going to change. And the world really didn’t change from Big Data, but I think AI is what Big Data should have been.”

Watch the show above. You can also hear Machine Meets World as a podcast, join the email list, and contact the show.

Image for post
Image for post
Photo by Ben Sweet on Unsplash

Audio + Transcript

Chris Drumgoole:
Even though the raw technologist will feel like the use of human oversight’s going to hold them back, in the reality it’s going to make the technology adoption go faster because it’s going to get people more comfortable with it . They’re going to understand it better. And then they’re going to start to trust it.

James Kotecki:
This is Machine Meets World, Infinia ML’s ongoing conversation about artificial intelligence. And my guest today is the CIO of DXC Technology, Chris Drumgoole. Welcome to the show.

Chris Drumgoole:
Hi, James. Thanks for having me.

James Kotecki:
DXC is a Fortune 500 company. Actually, I looked it up, you’re in the Fortune 125, 130,000+ employees. And your core business is IT services. And you are the Chief Information Officer. Given all the things that you probably have on your mind, how much time do you spend thinking about artificial intelligence?

Chris Drumgoole:
It’s a great question because when you do have a job like mine and everyone who has my job, you spend a lot of your time dealing with the plumbing. That’s how you keep the place running every day. But you’ve got to carve off a good amount of your time and thinking to really pay attention to what’s coming next, otherwise you just become a plumber. So I wouldn’t say AI is top of my mind every single day, but as I look at some of the bigger business problems we have and some of the opportunities to fix things, it tends to come up more and more often.

Chris Drumgoole:
I do think in the seat where I come from, a lot of people blur ML and AI. When I think about AI is really when I think about the technology making the jump from doing what we tell it to do better and better every day into starting to think about what it wants to do on its own. And that’s my layman’s definition. I don’t know if it’s the official one. Good ML is really just taking the instruction we gave it and refining that instruction to make it better and better and better every day. Whereas AI is actually coming up with insights and things that we may not have thought to ask the question in the first place .

James Kotecki:
How do you think about staffing something like AI? And what do you tell your team directionally about what they should be thinking about?

Chris Drumgoole:
I’m not a big fan of creating dedicated tiger teams because I think they tend to become a science project-y. So the approach we’ve taken is more, let’s find some good individuals who have the skills, and then let’s seed them out into teams that are actually doing real hard work. So for example, in our world, we collect all our data. We’re trying to get better insights about how to run our business better. That’s a great place to seed a few folks who really understand the technology, know how it works and make incremental progress in terms of making it better.

James Kotecki:
What’s a cool AI application, or even demonstration, that you’ve seen recently?

Chris Drumgoole:
Using AI to uncover trends in terms of employee behavior, not in a Big Brother-ish way at all, but in a, “Hey everybody’s home, everybody’s working all the time. How can we look for behaviors that people maybe don’t even realize they’re doing themselves and help them?” So that’s very nascent, but it’s something I saw recently that I thought was really relevant in this post-COVID world to help someone maybe take care of themselves a little bit.

James Kotecki:
So you’re a business obviously, you’re at the C-suite of a business. Where do you actually see it practically being applied to have the most business impact right now?

Chris Drumgoole:
I think it’s going to sound a little cliche because for the last God-knows-how-many years, everybody was talking about Big Data and everyone — and the world is going to change. And the world really didn’t change from Big Data, but I think AI is what Big Data should have been when everyone talks about it. So I think about, sitting where I am, and we’re collecting operational data on millions and millions of devices and servers all around the world, just trying to deliver a great IT experience.

Chris Drumgoole:
Our ability as a business to take that data and make a decision faster and better before a problem arises itself, transforms directly to how we make money and keep our customers happy. The ability to make even the smallest improvement in terms of an operating procedure, can have dramatic downstream impact. And I think that’s the promise of it from where we sit. Not there yet, and we haven’t really seen delivery of it — there’s some interesting places where there’s bright spots, green shoots, if you will.

James Kotecki:
When the behind-the-scenes AI is working, it almost doesn’t feel like it’s working. Or it just feels like things are working a little bit better or just more as they should, as I would have already expected them to.

Chris Drumgoole:
I think that’s spot on. The goal is for that little moment of surprise or joy where it’s like, “Oh, look! That’s there already,” or, “That’s done already.” That little ability to have, in our case, 130,000 people have to do one less thing a day has real impact. It’s decidedly not the cool, trendy technology that someone’s going to be up on stage, but our shareholders like it.

James Kotecki:
Let’s talk about AI ethics for a moment. How do you think about that? How do you define that challenge within DXC?

Chris Drumgoole:
To be clear, everything I’m talking about in terms of DXC itself is really around where we want to go versus where we are now, so very aspirational still. But I think it’s a really important question. There’s two pieces to the ethics conversation around AI. One is, really just doing what you should be doing already well, in terms of making sure that you’re working with your teams all over the world to make sure that they understand the technology that you’re using, they understand the data you’re collecting and they know how you’re going to use it and they’re comfortable with that because communicating that is half the battle. And I think a lot of people fall down on that. They just assume everyone understands the technology. They assume everyone understands the intent and then they don’t take the right actions.

Chris Drumgoole:
So I think that it sounds very base, but that’s really first, is treat it like any other technology in terms of really working with your people, however they’re organized around the world. The other part, which is I think where everybody goes to as AI ethics, is eventually how much control do you give the thing? And I think that gets really interesting and maybe I’m a little old school, but I think for now we have to build these things where there’s still quite a bit of final human oversight in the process in a way that’ll probably, to raw technologists, feel arduous and feel like we’re holding back the technologies for a while, because the reality is we are. Like any new technology, we don’t understand what it can fully do.

Chris Drumgoole:
If you look at the history of technology, literally going back to lighting and electricity, no one understood what it could do. And there were good effects and there were bad effects. But the speed of pace of the world meant that the bad effects were kind of quite limited and they didn’t affect the whole world. In the modern connected world, a bad piece of AI could make a really bad impact, really wide and really fast. So I think a little more human oversight while it learns is the way to go on that front.

James Kotecki:
So if humans continue to have oversight, is that something that you need to be upfront and communicating to your employees, maybe to allay fears that the people might have or concerns about AI and automation and machine learning coming in and disrupting the work that they’re doing?

Chris Drumgoole:
Yeah, I think it’s super important because again, I think technology is only good as the people who accept the technology. To get faster acceptance — again, even though the raw technologist will feel like the use of human oversight’s going to hold them back, in the reality it’s going to make the technology adoption go faster because it’s going to get people more comfortable with it. They’re going to understand it better. And then they’re going to start to trust it. And the best way for the oversight to go away is the day that you don’t even realize you made a conscious decision to go away. Everybody just got comfortable enough that now this is how it is.

James Kotecki:
There may still be a need for some oversight, maybe not as much, even after that happens, because you could have, “Oh, it worked a hundred days out of a hundred. It was fine.” But maybe there’s an edge case on that hundred and first day that we didn’t realize where we actually should have still had some kind of eyes on it or some kind of technology to oversee it.

Chris Drumgoole:
Totally. And I think, listen, if nothing else, the last eight months of life and — COVID is the human race’s edge case. Hopefully, among with the many, many bad things that have happened as a result of this pandemic, hopefully one of the good things is maybe we as a society have a little more appreciation for what we previously would have thought of as impossible events. And that very much plays into AI. Just because it worked a hundred days, doesn’t mean it works on a hundred and one. As the AI starts to improve itself is when the human oversight really becomes important because the mission starts to creep.

James Kotecki:
Speaking of mission, our last question. I did some research on you. I found out on LinkedIn, you were a firefighter/EMT for many years. That’s a pretty intense set of experiences. Did that experience — does that experience shape the way you think about AI, ML or just technology generally?

Chris Drumgoole:
It definitely shaped the way I think about operating technology. My bad joke with my employees is I’m still fighting fires. They’re just not chemical anymore. I think it totally does. Amongst many things, the fire service and the military service do a great job of prioritization. Raw, brutal, just what’s important now at the expense of everything else and how to achieve that mission amongst all others, because that’s the nature of their business, right? Save the life, put out the fire, take the hill. I think that as you manage technology, that’s a really valuable skillset. Some of my people thought I was crazy when I brought in the FDNY to teach a bunch of technologists about incident management, but you know who knows how to do it really well? The FDNY. So it doesn’t matter what kind of incident. I think there’s a lot of appropriate stuff there.

James Kotecki:
Chris Drumgoole, the CIO at DXC. Thanks so much for joining us today on Machine Meets World.

Chris Drumgoole:
Thanks a lot, James. Take care.

James Kotecki:
And thank you so much for watching. You can email the show, mmw@infiniaml.com. Please like us. Share this. You know what to do. I am James Kotecki and that is what happens when Machine Meets World.

Machine Meets World from Infinia ML

Weekly Interviews with AI Leaders

Medium is an open platform where 170 million readers come to find insightful and dynamic thinking. Here, expert and undiscovered voices alike dive into the heart of any topic and bring new ideas to the surface. Learn more

Follow the writers, publications, and topics that matter to you, and you’ll see them on your homepage and in your inbox. Explore

If you have a story to tell, knowledge to share, or a perspective to offer — welcome home. It’s easy and free to post your thinking on any topic. Write on Medium

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store