On this episode of The Six Five In the Booth, hosts Daniel Newman and Patrick Moorhead welcome Andrew Coward, General Manager, Software Networking at IBM and Mark McDonald, VP, Wireless Networks at Bell Canada at MWC 2024 for a conversation on how generative AI is being used in the networking landscape.
Their discussion covers:
- The current state of wireless networking and what issues can be solved with AI
- How generative AI is currently used in networking and how will this change over time as the use of generative AI matures
- Thoughts from Andrew and Mark on how generative AI will transform both of their respective companies
Click here to learn more about IBM’s announcements at MWC.
Be sure to subscribe to The Six Five Webcast, so you never miss an episode.
Watch the video here:
Or Listen to the full audio here:
Disclaimer: The Six Five webcast is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors and we ask that you do not treat us as such.
Transcript:
Patrick Moorhead: The Six Five is on the road in Barcelona, Spain for Mobile World Congress 2024. We are here in the IBM booth. Dan, we can just say it, I’m going to call it here. Mobile World Congress is back. There are tons of people in the hallways, chatting, doing business deals, pontificating, talking about the future, talking about value add today.
Daniel Newman: Yeah, Pat. This show has always been one of the most instrumental. Anytime that you think about the importance that the network plays in your life, you just pull out this device, whatever device you’re on, and you think about, would I be able to do whatever I’m doing right now if I didn’t have the network? Yes, sure, maybe, sometimes you’re on the Wi-Fi, but that’s a network too, by the way.
But that carrier network, what the service providers do to enable what we’re doing with generative AI, AI, machine learning, running our applications, social media with our friends, and of course, all this cool new stuff we’re seeing with augmented and virtual reality, Pat, this is powerful stuff. Sometimes I think we take it for granted a little bit, but yeah, I mean 95,000 people here, 95,000. This show is almost back to the full capacity. Boy, have I seen a lot of great stuff.
Patrick Moorhead: By the way, it only took us a minute and a half to get to the word AI. That is permeating not only the-
Daniel Newman: I took my time.
Patrick Moorhead: … consumer play, but also B2B. It’s an enabler. It’s an enabler of business, it’s an enabler of value. By the way, it’s really hard and we still haven’t cracked it yet, but we do have two guests that are working very hard to figure all of this out. Andrew, IBM, nice to see you again. Usually, we say first time guest, but second time guest and two years in a row. Of course, Mark from Bell Canada. Great to see you.
Mark McDonald: Great to see you as well.
Patrick Moorhead: Absolutely.
Daniel Newman: All right. Gentlemen, you heard my preamble. I talked a little bit about what’s going on in the state of the network. I had a great couple of meetings and we were actually talking about how sometimes it’s so easy to forget, lose the appreciation for all that these companies here, these service providers do to enable us every day on every one of our devices to be able to do all of the important things that we do, whether it’s work or play, connecting with our family and our friends. But let’s talk about wireless networks a little bit. Andrew, I’m going to start this one off with you, but maybe you talk a little bit about what you see the state of wireless networks today, in particular with AI. What’s the problem that we’re coming together here to solve?
Andrew Coward: Well, to your point, right, the deployment of 5G networks, the deployment and the expectations that we have of mobile networks are absolutely huge. The promise of AI, and we’ll talk about whether we’re going to get there or not and how quickly, it’s really that it massively can help reduce the operational cost of the network, which is the headlines that the Wall Street Journal gets. But more importantly, it’s going to massively speed up how quickly it takes to run a process, how easy it is to solve customer challenges, customer problems, how much effort goes into turning on a new subscriber or solving a problem within a network. The time compression that’s going to happen will greatly increase the efficiency, which leads to the cost dynamic.
Daniel Newman: Mark, what’s the perspective from where you sit?
Mark McDonald: Yeah. Maybe I can give you the context what’s going on in Canada? In wireless networks, the competitive landscape has, I would say, changed dramatically in recent times in Canada. I’ll just give you an example. In the past 12 months, 2023, the average price that a consumer pays for data from mobile broadband has reduced by 27% just in one year. That’s against the backdrop of consumer price inflation of plus 4%.
We’re in a real business problem here. The costs of the network, of course, are going in the other direction. The cost of the network are in fact increasing. We have to solve this business contextual problem, and that’s where gen AI and AI will play a role going forward. Things like how do we more intelligently decide on where to invest in the network because of that business challenge, how do we build and operate the networks more cost effectively, and arguably most important, how do we use generative AI and AI to get a better customer experience across our networks.
Patrick Moorhead: Generative AI is relatively new. Machine learning algorithms go back to the ’60s, and we really saw actually it flourished, started University of Toronto using GPUs to accelerate visual object recognition algorithms. But where are we? What’s the maturity level now? I think I might know the answer to this, but also want to caveat that machine learning is nothing new to the network. It’s very much alive and well. Where are we on the map right now?
Mark McDonald: As it pertains to generative AI on wireless networks, I’d say it’s totally nascent. I think we’re in the very, very early stages-
Patrick Moorhead: It’s in the research phase, beyond the research phase?
Mark McDonald: I would say it’s in the experimentation phase. I would say there is a very high cadence of the technology availability becoming available. Therefore, there’s a great interest in how do we utilize this to solve problems that we have. The speed of adoption seems to be fast, but to be honest, the business value outcomes as a result of the experimentation are not proven out fully. Maybe I can help a little bit. You mentioned AI, and some people are calling it classic AI or traditional AI, but even though it’s been existing for a long time, only in recent years has it become very valuable in wireless networks.
One of the examples I can give is for customer experience. We’ve been using predictive analytics to really get down to a single customer of one and what their experience is like on the network, and normally detection for customer degradations. That’s positive. Some of the experimentation with gen AI is how do you build upon that. How do you use gen AI models to be able to look for root cause analysis? How do you do it for resolution? How do you look for categorization of the customer issues? So, the interest is high on what it could do, but the realization is very early.
Andrew Coward: I think there’s a question really about the readiness of telcos on whether they’ve got the right data and if they have the right data, do they actually have the ability to orchestrate the network and actually take action. It’s really simple. I mean, if you imagine a query that would come in like why was Andrew’s social media experience terrible yesterday? He couldn’t post. Do we have the right data to be able to answer that natural language query? And then further think about a query or a command to increase the capacity to FC Barcelona’s football stadium and all of the surrounding public transport systems by another 30,000 spectators.
Could the network actually do that, or would it take six weeks to go to provision that? The danger with AI in that context is that if we apply it too soon, it’s going to be fairly toothless because it’s just going to be able to answer why was my bill $20 more expensive this month. That’s a perfect use case for where we are today, but these more deep use cases of where we have to get to the savings and get to the customer value.
Daniel Newman: Yeah. That’s a problem that could be handled by generative in many ways. Why is my social media performing so poorly? It’s, A, the network is bad, or B, it was your post was bad and generative AI decided not to allow you to actually put that online. Maybe we’ll get there at some point. I’m just having a little fun with you. Mark, you made a comment about right data. Bell Canada, for instance, in order to do this, it is about having the right data, it’s having the right data, and of course, there’s a lot of ethics, there’s governance, something IBM is, by the way, very, very focused on, and it did with watson.governance very early on.
But how are you thinking about that part of the right data, making sure that people are getting a great experience, but at the same time trusting that they still are having levels of privacy, that their data is not being abused for things like advertising. I mean, it’s a balance. Where are you at with that?
Mark McDonald: First of all, we have a lot of data. I think your point is right, we have the right data, and I think it’s a problem about data organization, the organization of the data across the various areas that lies in the network. A big part of it’s about the reliability of the data. It’s about the location of the data, it’s about the availability of the data, it’s about the right access to the data with the right controls that you’re referenced based on the user that wants to access it. Is that the right data they should be able to access?
In Bell, we feel this as a foundation to the success of gen AI, is having that data organization. We have a program we call DaaS, data as a service, where we’re using a mesh architecture across our various data sources with all of the right connectors that will plug into those various sources using the power of public cloud to be able to help accelerate some of that adoption. With that foundation, we think we’ll be able to get some of those governance, security, privacy, and quality data integrity elements correctly organized.
Patrick Moorhead: My first computer class that I had in high school, they taught me garbage in, garbage out, which very much is what we’re talking about. Out of one side of my mouth, I’m going to say what’s old is new. We’ve been here, done that. But as I, I don’t know, take a more mature view, generative AI is different, because one of the elements that it is different, it enables co-mingling of different types of data. ERP, CRM, product line management, HRM, and all these. That’s the possibility. Machine learning was really narrowed in on a certain subset of type of data and typically it didn’t get out of its swim lane. So, having a data management strategy upfront is an absolute requirement, particularly if you’re going to be co-mingling these data sets.
When we talk to different types of businesses, that is where the big, say, home runs in American context are going to be, and it’s a super exciting moment. I’m actually getting to a question here. The two of you have some incredible history, I mean, hundreds of years of doing this. Obviously, I’ve got two network experts up here and practitioner and vendor who helps you do these things. What areas are you looking at using generative AI outside of the network, potentially co-mingling these types of data?
Mark McDonald: That’s a great question. Yeah, Bell will be 134 years old in April of this year. We’ve gone through multiple transformations over our history and certainly gen AI is one of those that we feel will really help our digital transformation in the coming years and across the organization more broadly, a lot of experimentation as well on those value adding use cases. Couple of simple examples. Knowledge management is a great one. Stuff like code assistant. We have a lot of software developers that there is a peer review process on code before we push it into production. The ability to be able to use generative AI to take out some of that monotony of code reviews versus the creativity of the code is quite interesting. We’re doing some experimentation around that.
There’s other stuff like documentation, be it in the network for our vendors, very complex documentation, how to navigate through that in a very simple way to be able to get proposals and answers very quickly. But also on things like employee services across the corporation. My benefits, my company policies, so making that very consumable and easy to digest. On the business side, lots of opportunity to use these huge amounts of data for innovation and products and services for our various networks.
There’s an opportunity really to be creative, to look at network trends, to be able to predict what those trends mean and have very customizable offers and products and services for our customers. And then, ultimately, customer experience. End to end is going to be a critical use case that’s very, very personalized, that’s proactive, that’s self-serve, fully digital, using all those techniques, like chatbots and other methods. Yeah, we got excited about what the future holds.
Patrick Moorhead: Yeah, the three to four areas that you hit are absolute goldmine opportunities, and you won’t be the only one doing them, which I think is good, because quite frankly, when it comes to a carrier, you have very little room for error in what you do. IBM has been very aggressive and targeted three areas that you’re serving your clients, but you’re using it yourself. What are some of the areas inside, I mean, Arvind has been very communicative on what you’re doing inside the company.
Andrew Coward: Well, right, I think it’s called eating your own dog food, although I’m not sure I like that term.
Patrick Moorhead: You still never found a better one.
Daniel Newman: Drinking your own champagne.
Andrew Coward: Drinking your own champagne. I like it. Let’s use that.
Daniel Newman: There you go.
Andrew Coward: No, that’s obviously right. I think the data part of that conversation, when we released what’s next on data, so much of that was about how we consolidate different data sources from different places, completely disparate.
Patrick Moorhead: With different security levels, by the way.
Andrew Coward: Different security levels and different. Some of it public, some of it within your own system. We applied that internally to our HR systems with our HR chatbot. It wasn’t toothless. It could answer pretty much any question about how are any of our employees, what our policies are, how much we going to get paid, when their holidays were, all those kinds of things. Pretty powerful things. But the power of association opens up a whole load of new use cases, things that people aren’t necessarily thinking about.
For example, if you are a quick serve restaurant, how narrow are all the distribution centers for drinks or for food with suppliers. Those aren’t things that you would have necessarily, you wouldn’t have them in your own database and node structure. You might have obviously where your restaurants are. Being able to correlate those kinds of data sets becomes really interesting just by simple query. The union of all SQL requests, if you like, are going to come together to. Obviously, it’s not SQL these days, but you get the point. It is that union, which I think is really valuable, that will enable that types of queries that people have been thinking about that they go, “Oh, I wonder if I could find that out.” I think that’s what’s really fascinating about their own data sets and their own systems.
The other thing we did with our platform, of course, was we enabled our customers to use any of the available language model, LLMs, that they wanted to on the same dataset. Pick your own journey as to what you get the best results for, and I think that’s been really valuable in terms of the open ecosystem that we’re creating for that. But it goes back to, you talked about the age of IBM, and I think once you get over 100, you stop talking about how old you are, perhaps I dont know. But since the founding of IBM, it’s always been about accelerating business. From the first adding machines to the mainframes. This is just that, in simple stuff, that next acceleration of business, the next level of efficiency, next level of automation that we’re getting to. That’s why it’s so exciting. It’s that time of revolution that we’ve had, maybe only three or four times through the history of the company.
Daniel Newman: Well, Mark and Andrew, clearly the network is transformative. Of course, we are in the next wave of growth with AI. That is going to change every single industry. I’ll finish where I started that without connectivity, without a strong network and of course without strong wireless, a lot of what we’re doing even right here, right now, the information, the real time, the connectivity, the fact that businesses keep running while all of us are here is enabled by the power of the network, and only getting more exciting and powerful with generative AI. Mark, Andrew, I want to thank you both for joining us here on The Six Five. We look forward to tracking your journey and talking with you again soon.
Andrew Coward: It’s been a great pleasure.
Patrick Moorhead: Thank you.
Mark McDonald: Thank you.
Daniel Newman: All right, everybody. Hit that subscribe button. Join us for all the coverage of The Six Five here at MWC 2024. We appreciate you joining us. We hope you come back soon. We’ll see you all later.
Author Information
Daniel is the CEO of The Futurum Group. Living his life at the intersection of people and technology, Daniel works with the world’s largest technology brands exploring Digital Transformation and how it is influencing the enterprise.
From the leading edge of AI to global technology policy, Daniel makes the connections between business, people and tech that are required for companies to benefit most from their technology investments. Daniel is a top 5 globally ranked industry analyst and his ideas are regularly cited or shared in television appearances by CNBC, Bloomberg, Wall Street Journal and hundreds of other sites around the world.
A 7x Best-Selling Author including his most recent book “Human/Machine.” Daniel is also a Forbes and MarketWatch (Dow Jones) contributor.
An MBA and Former Graduate Adjunct Faculty, Daniel is an Austin Texas transplant after 40 years in Chicago. His speaking takes him around the world each year as he shares his vision of the role technology will play in our future.
