 from downtown San Francisco. It's theCUBE, covering IBM Chief Data Officer Strategy Summit 2018, brought to you by IBM. Welcome back to San Francisco, everybody. You're watching theCUBE, the leader in live tech coverage. My name is Dave Vellante, and we are here at the IBM CDO Strategy Summit, hashtag IBM CDO. Caitlin Halferty is here. She's a client engagement executive for the Chief Data Officer at IBM. Caitlin, great to see you again. Great to be here, thank you. And she's joined by Brandon Purcell, who's a principal analyst at Forrester Research. Good to have you on. Thanks very much, thanks for having me. First time in theCUBE, you're very welcome. I'm a newbie. Caitlin, that's right, you're a CUBE alum in no time, I promise you. So Caitlin, let's start with you. This is, you've done a number of these CDO events. You do some in Boston, you do some in San Francisco, and it's really great to see the practitioners here. You guys are bringing guys like Interpol to the table, you've announced your blueprint, and the audience seems to be lapping up the knowledge transfer. So what's the purpose of these events? How has it evolved, and just set the table for us? Sure, so we started back in 2014 with our first Chief Data Officer Summit, and we held that here in San Francisco. Small group, probably had about 30 or 40 attendees, and we said, let's make this community focused peer-to-peer networking. We're all trying to build the role of either the Chief Data Officer or whomever's responsible for enterprise-wide data strategy for their company, a variety of different titles. And we've grown that event over the, you know, since 2014, we do Spring in San Francisco, which tends to be a bit more on the technical side, given where we are here in San Francisco and Silicon Valley, and then we do our business-focused sessions in fall in Boston. And I have to say, it's been really nice to see the community grow from a small set of attendees, and now we're at about 130 that join us on each coast. So we've built a community in total of about 500 CDOs and data executives that are with us on this journey, so that's been great. And Brandon, your focus at Forrester, part of it is AI. I know you do some other things in analytics, the ethics of AI, which we're going to talk about. I have to ask you, from Forrester's perspective, we feel like we're entering this new era of, you know, there's digital, there's data, there's AI. They seem to all overlap. What's your point of view on all this? So I'm extremely optimistic about the future of AI. I realize that the term artificial intelligence is incredibly hyped right now, but I think it will ultimately fulfill its promise. If you think about the lifecycle of analytics, analytics start their lives as customer data, as customers interact and transact with you, that creates a footprint that you then have to analyze to unleash some sort of insight that customers likely to buy or churn or belongs to a specific segment. Then you have to take action. The buzzwords of the past have really focused on one piece of that lifecycle, big data, the data piece. Not much value unless you analyze that, so then predictive analytics machine learning. What AI promises to do is to synthesize all of those pieces from data to insights to action and continuously learn and optimize. It's interesting you talk about that in terms of customer churn. I mean, with the internet, there was like a shift in the balance of power to the consumer, but it used to be that the brand had all the knowledge about the buyer and then with the internet, we shop around, we walk into a store and look at them, we go buy it on the internet, right? Now that AI maybe brings back more balance, symmetry. I mean, what are your thoughts on that? Are there clients that you work with trying to regain that advantage so they can better understand the customer? Yeah, that's a great question. I mean, if there's one kind of central ethos to Forester's research is that we live in the age of the customer and understanding and anticipating customer needs is paramount to be able to compete, right? And so it's the businesses in the age of AI and the age of the customer that have the data on the customer and the ability to distill that into insights that will ultimately succeed. And so the companies that have been able to identify the right value exchange with consumers to give us a sense of convenience so that we're willing to give up enough personal data to satisfy that convenience are the ones that I think are doing well. And certainly Netflix and Amazon come to mind there. For sure, and now of course that gets into the privacy and the ethics of AI, IBM's making a big deal out of this. You own your data, you're not trying to monetize, figure out which ad to click on. Maybe give us your perspective, Caitlin, on IBM's point of view there. Sure, so we lead with this thought around trust in your data. Your data is your data. Insights derive from that data, your insights. We spend a lot of time with our Watson legal folks and one of the things, pieces of material we've released today is the real detail at every level, how you engage the traceability of where your data is so you have a sense of confidence that you know how it's treated, how it's curated. If it's used in some third party fashion, the ability to know that, have this ability into it, the opt-out, opt-in opt-out set of choices, making sure that we're not exploiting the network effect where perhaps party C benefits from data exchange between A and B, that A and B do not, or do not have an opportunity to influence. And so what we wanted to do here at the summit over the next couple of days is really share that in detail and our thoughts around it and it comes back to trust and being able to have that visibility and traceability of your data through the value chain. So of course, Brandon, as a customer, I'm paying IBM, so I would expect that IBM would look out for my privacy and make that promise. I don't really pay Facebook, right? But I get some value out of it. So what are the ethics of that? Is it a pay or no pay, or is it a value or no value? Or is it everybody really needs to play by the same rules? How do you parse all that? You know, I hate to use a vague term, but it's a reasonable expectation. Like I think that when a person interacts with Facebook, there's a reasonable expectation that they're not going to take that data and sell it or monetize it to some third party like Cambridge Analytica. And that's where they drop the ball in that case. But that's just in the actual data collection itself. There's also, there are also inherent ethical issues in how the data is actually transformed and analyzed. So just because you don't have like specific characteristics or attributes in data like race and gender and age and socioeconomic status, in a multi-dimensional data set, there are proxies for those through something called redundant encoding. So even if you don't want to use those factors to make decisions, you have to be very careful because they're probably in there anyway. And so you need to really think about what are your values as a brand and when can you actually differentiate treatment based on different attributes? Because you can make accurate inferences from that. Yeah, you absolutely can. Is it the case of actually acting on that data or actually the ability to act on that data? If that makes sense to you. In other words, if an organization has that data and could in theory make the inference but doesn't, is that crossing the line? Is it the responsibility of the organization to identify those exposures and make sure that they cannot be inferred? Yeah, I think it is. I think that that's incumbent upon organizations today. Eventually regulators are going to get around to writing rules around this and there's already some going into effective force in Europe with GDPR at the end of this month. But regulators are usually slow to catch up. So for now it's going to have to be organizations that think about this and think about, okay, when is it okay to treat different customers differently? Because if we break that promise, customers are going to ultimately leave us. That's a hard problem. Right, right. You guys have a lot of these discussions internally and can you share those with us? Yeah, absolutely, we do. And we get a lot of questions. We often engage with the data strategies perspective and it starts with, hey, we've got great activity occurring in our business units and our functional areas but we don't really have a handle on the enterprise wide data strategy. And at that point, we start talking about trust and privacy and security and what is your data flows look like? So it starts at that initial data strategy discussion and one other thing I mentioned in my opening remarks this morning is, we released this blueprint and it's intended, as you said, to put a framework and process and reflect a lot of the lessons learned that we're all going through. I know you mentioned that many companies are looking at AI adoption, perhaps more so than we realize. And so the framework was intended to help accelerate that process. And then our big announcement today has been around the showcases, in particular our platform showcase. So it's really the platform we've built within our organization, the components, the products, the capabilities that drives for us. And then with the intent of hopefully being illustrative and helpful to clients that are looking to build similar capabilities. So let's talk about adoption. Yeah, sure. You often hear this bromide, we live in a world where the pace of change is so fast and things are changing so quickly, it's hard to deny that. But then when you look at adoption of some of the big themes in our time, whether it's big data or AI, digital, block change, there are some major barriers to adoption. So you see them adopted in pockets. What's your perspective and Forrester's perspective on adoption of, let's call it machine intelligence? Yeah, sure. So I mean, every year Forrester does a global survey of business and technology decision leaders called business technographics. And we ask folks about adoption rates of certain technologies. And so when it comes to AI globally, 52% of companies have adopted AI in some way and another 20% plan to in the next 12 months. What's interesting to me actually is when you break that down geographically, the highest adoption rate, 60 plus percent, is in APAC, followed by North America, followed by Europe. And when you think about the privacy regulations in each of those geographies, well, they're far fewer in APAC than there are and will be in Europe. And that's, I think, kind of hamstringing adoption in that geography. Now, is that a problem for Europe? I don't think so actually. I think AI, the way AI is going to be adopted in Europe is going to be more refined and respectful of customers' intrinsic right to privacy. You know, I want, go ahead. I have to say, Dave, I have to put a plug in. I've been a huge fan of Brandon's for a long time. I've actually, you know, a few years now of his research and some of the research that you're mentioning, I hope people are reading it because we find these reports to be really helpful to understand, as you said, the specifics of adoptions, the trends. So I've got to put a plug in there because, you know, the quality of the work and the insights are incredible. So that was why I was quite excited when Brandon accepted our offer to join us here in this session. Yeah, so let's dig into that a little bit. So it seems like, so 52%, I'm wondering what the other 48 are doing. They probably are, and they just don't know it. So it's possible that the study looks at, you know, a strategy to adopt, presumably. I mean, actively adopting, but it seems, I wonder if I could run this by you to get your comment. It seems that people will, organizations will more likely be buying AI as embedded in applications or systems or just kind of invisible than they will necessarily be building it. I mean, many are trying to probably build it today and what's your thought on that in terms of just AI, you know, infused everywhere? So the first foray for most enterprises into this world of AI is chatbots for customer service. I mean, we get a ton of inquiries and forester about that and there are a number of solutions. IBM certainly has one that fulfill that need and that's a very narrow use case, right? And it's also a value additive use case. If you can take, you know, more of those call center agents out of the loop or at least accelerate or make them better at their jobs, then you're going to see efficiency gains. But this isn't this company-wide AI transformation. It's just one very narrow use case. And usually that's, you know, most elements of that are pre-built. We talked this morning, or the speakers this morning talked about commoditization of certain aspects of machine learning and AI. And it's very true. I mean, you know, machine learning algorithms, many of them have been around for a long time and you can access them for multiple different platforms. Even natural language processing, which a few years ago was highly inaccurate, is getting really, really accurate. So when in a world where all of these things are commoditized, it's going to end up being how you implement them that's going to drive differentiation. And so I don't think there's any problem with buying solutions that have been pre-built. You just have to be very thoughtful about how you use them to ultimately make decisions that impact the customer experience. I want to, in the time we have remaining, I want to get into the tech radar, the sort of taxonomy of AI or machine intelligence. You've done some work here. How do you describe, can you paint a picture for what that taxonomy looks like? So I think most people watching realize AI is not one specific thing, right? It's a bunch of components, technologies that stitch together lead to something that can emulate certain things that humans do, like sense the world around us, see, read, hear, that can think or reason, that's the machine learning piece, and that can then take action, and that's the kind of automation piece. And there are different core technologies that make up each of those faculties. The kind of emerging ones are deep learning. Of course, you hear about it all the time. Deep learning is inherently the use of artificial neural networks. Usually to take some unstructured data, let's say pictures of cats and identify this is actually a cat, right? Who would have thought that would have led to the spoon, right? Right, exactly. That was something you couldn't do five or six years ago, right? You couldn't actually analyze picture data like you analyze row and column data. So that's leading to a transformation. The problem there is that not a lot of people have this massive number of pictures of cats that are consistently and accurately labeled cat, not cat, cat, not cat, and that's what you need to make that viable. So a lot of vendors, and Watson has an API for this, have already trained a deep neural network to do that so the enterprises aren't starting from scratch. And I think we'll see more and more of these kind of pre-trained solutions and companies gravitating towards the pre-trained solutions and looking for differentiation, not in the solutions themselves, but again, how they actually implement it to impact the customer experience. Well, that's interesting. This is hearing you sense, see, read, hear, reason, act. These are words that describe not the past era. This is a new era that we're into. We're in the cloud era now. We could sort of all agree with that, but the cloud doesn't do these things. We are clearly entering a new wave. Maybe it's driven by Watson's Law or we'll see whatever holds out. All right, Caitlin, I'll give you the last word. Put a bumper sticker on this event and where we're at here in 2018. I'll say it's interesting to watch the themes evolve over the last few years. We started with sort of a defensive posture. Most of our data executives were coming, perhaps from an IT type background. We see a lot more with line of business in a chief operations type role. And we've seen the, we still, King of the Data Warehouse, sort of how we described at the time. And now I see our data leaders really driving transformation. They're responsible for both the data as well as the digital transformation. On the data side, it's the AI focus and trying to really understand the deep learning capabilities, machine learning that they're bringing to bear. So it's been, for me, it's been really interesting to see the topics evolve, see the role and the strategic piece of it as well as see these guys elevated in terms of influence within their organization. And then our big topic this year was around AI and understanding it. And so having Brandon share his expertise was very exciting for me because he's our lead analyst in the AI space. And that's what our attendees are telling us. They want to better understand and better understand how to take action to implement and see those business results. So I think we're going to continue to see more of that. And yeah, it's been great to see, great to see it evolve. Well, congratulations on taking the lead. This is a really important space. A lot of people didn't really believe in it early on. Thought the chief data officer role would just sort of disappear. But you guys, I think, made the right investment and a good call. I was laughed out of the room when I proposed. I said, hey, we're hearing of this doing a market scan of chief data officer either by title or something similar, title responsible for enterprise wide data. I was laughed out of the room. I said, let me do a qualitative piece. Let me interview 20 and just show. And then you're right. It was the thought was the role is going to go by the wayside. And I think we've seen the opposite. Oh yeah, absolutely. Data has grown in importance. The associated capabilities have grown. And I'm seeing these individuals, their scope of spear of responsibility really grow quite a bit. The forces track this. I mean, you guys, I think just a few years ago, it was like, oh yeah, 20% of organizations have a chief data officer. Now it's much, much higher. Yeah, yeah, it's approaching 50%. Yeah, so, yeah. All right, Brandon, Katelyn, thanks very much. Thanks for having us. Thank you, it was great, great. Keep it right there, everybody, we'll be back at the IBM Chief Data Officer Strategy Summit. You're watching theCUBE.