 From around the globe, it's theCUBE with digital coverage of IBM Think 2021, brought to you by IBM. Welcome back everyone to the CUBE coverage of IBM Think 2021. I'm John Furrier, host of theCUBE. We've got a great guest here, Scott Hebner, Vice President of Marketing at IBM for data and AI. CUBE alumni has been around the wave around data, had many conversations over the years. Scott, welcome back to theCUBE. And I wish we were in person, but we're remote for the virtual conference of Think 2021. Thanks for coming on. Hey John, great to be here. And yeah, I guess we have adapted to the world of being on a screen all day long. Well, great to have you in one of the things about virtualization of media is that we get more content. This year, there's so many more signature stories around IBM Think. And one of the things that's really fun for us is the data conversations and AI as the transformation and innovation equations are coming together at scale. You're seeing an accelerated piece here. My first question for you is this digital shift that's going on, the preferences are shifting to virtual now digital in the wake of COVID. What do companies need to adapt from your perspective? As you see this playing out, what's your perspective? It's interesting to use that term because we've been calling it the great digital shift. And yeah, there was an interesting survey, a pretty big survey of global C-suite that McKinsey did. And they pointed out that 79% of those leaders felt that COVID highlighted the immaturity of their digital capability. And while they thought they were on the right path and they were building strong digital capabilities, the whole world of the pandemic, remote work, how you engage with customers, call centers going off the hooks in terms of people calling. It just goes on and on and on. And they also pointed out that 90, I think it was 96% of them are going to speed their digital reinvention. And you mentioned data, if you think about it, it's data that fuels digital capabilities, right? What could as digital if it's not data, right? It's all data. So it's the fuel that makes it all work. And when you think about the ability to leverage all your data, you got to democratize it because it's siloed all over the place. It's growing at six times rate over the next three years. It's really all over the place every touch point across the digital ecosystem. And the only way to deal with the data and to unlock its value, particularly in predictive ways is through AI, right? And so what we're seeing is a huge amount of investment in multi-cloud, really bringing together this notion of hybrid and then applying AI as the intelligence to create a more predictable and resilient business, right? Through a digital model, right? And it's really, the investment is really going through the roof. I think AI has been demystified over the years been a lot of people saw the machine learning and now you got NLP and data control planes that are making it more addressable. But the real thing that comes up here I think this year is this role between business and consumer and AI has that kind of dynamic. And I want to ask you because I was just having a conversation with one of your partner, IBM's partner, Samsung Casey Choi runs EVP for the B2B B2G group at Samsung. It's a huge IoT thing and AI is a big part of that consumer. We talked about the consumer electronics, business issues. How is AI different for business versus the consumers? I mean, obviously you got industrial IoT edge and you got automation piece. What's the difference? I mean, someone asked you that between business and consumer AI? Yeah, I mean, actually, I think that's one of the areas that we really differentiate ourselves and we're putting a bulk of investments is notion of AI for business, right? And a lot of people think of AI sometimes they think of Siri and Alexa and things that go on in your car and all that. Obviously that's a big part of applying machine learning and all that. But when we talk about AI for business, we're thinking about four core attributes. One is that it needs to understand the unique language of your business and industry, right? And that's not just natural language, but it's the ability to debate. It's the ability to read documents, interpret documents. It's the ability to really understand the context because you and I can ask the same question in five or six different ways. And it needs to understand the business to be able to interpret that and help answer the question. Unlike Siri or Alexa, where you really got to have the right semantics and it won't understand the nuances as well. So understand the language of business is one. Two is that we believe AI is the engine for automation. So AI is really about automating workflows and experiences. Because anything that you want to automate and make more productive, you have to have some predictive capabilities to it to understand what to do. You have to learn about what's trying to be accomplished, which is always unique and personalized. So that's the second one is about automation. The third is about driving trust in outcomes, right? In the business outcomes, which means if so many of our models say, Scott, go jump off a bridge, I probably wouldn't want to do that unless it really explained to me convincingly that I should do that and maybe I will. But explainability and trust is such a critical part of AI for business. And then finally, it needs to run everywhere. It has to integrate everything. And we believe, unlike a lot of the competitors where you have to bring the data to AI, we're saying leave the data where it lives and bring AI to the data. So it runs anywhere from the data center to the edge, the same model, the same capabilities in a distributed environment. So those four kind of attributes come together to what we call AI for business. And that's what's going to allow call centers and supply chains and business planning and risk and regulatory mitigation, those kind of things to really come to life in a predictive way. Without those attributes, it's much harder to do a lot more coding and you're not going to have as much accuracy. Yeah, I mean, what you're just walking through there is interesting. And if you think about consumer, okay, yeah, Alexa, go get me, you know, what's the weather like in Palo Alto or whatever, you know, those kinds of things, it's all back in pretty complicated, but it's not as complicated as moving data to the edge and moving computer around. And the complexity of dealing with data has always been an open discussion, but now with AI such at the center point of the value project and becoming table stakes. I mean, we're hearing companies say, if you don't have an AI innovation strategy, you're going to be irrelevant or even delisted from the stock market. That's some radical views, but talk about this complexity and how it's being tamed for customers because if you don't have the data exposed, you're only as good as the data that you have. And this has been a conversation we've had on theCUBE many times before with you and some of your other peers here at IBM. If you can't get the data, what good is it? The insights are only as good as what you can program. So this means that data's got to be accessible and it's also complexity to move it around. So can you unpack that equation? Yeah, it's the whole notion of garbage in garbage out and AI, you know, AI, it's lifeblood is data. And we have a quib that we always say that there's no AI without an IA, an information architecture. And we are well over 30,000 engagements among our clients around AI. You know, we have the AI ladder, which is our prescriptive approach. We've learned a ton over the years. And we said before, you know, the great digital shift, well, the great inhibitor is the complexity of all this data. And the average large enterprise has over 1,000 repositories and sources of data. As things go out into the edge, that's just going to multiply. There's more and more movement to put applications, you know, software as a service applications on the cloud and most businesses have multiple clouds. So you're further fragmenting all the data. And if you look at what like Gartner has said and many others, these big data projects of the past are very slow, they're costly and they've had limited impact. This idea of moving data, replicating data is just not going to work as the explosion of data increases in terms of touch points, in terms of types and just terms of pure velocity. And also at the same time, the value of data, it's lifespan is rapidly decreasing. A customer record that was created yesterday may not be as valuable a year from now or even in three months from now because things change so much, right? Yeah. All right, so I got to ask you the question then because this is kind of, I'm a customer. What's in it for me? End of the day, I got data problem, you got my attention. I got to move data, I got the edge, hybrid cloud has been defined as bona fide, it's done deals, hybrid multiclouds around the corner, but that's just a subsystem of the operating system that's business now. So hybrid cloud is the operating model. Data is super critical. What does IBM offer? What can you offer me as a customer and why is it good? You guys got some announcements with CloudPak for data specifically here at Think. What's the solution? How do I solve this? What's IBM offering? Yeah, so I think it starts with the fact that we have a fully unified data and AI platform, meaning that they're not separate thoughts, they're all unified together as one life cycle and it runs anywhere on any cloud data center to the edge. So it starts with that notion, it helps you collect, organize and analyze data and then use AI throughout the business. Now when it comes to the data complexity, three core principles that we're putting to the next version of CloudPak for data. One is automation is inevitable. It's the only way to deal with all this complexity. Leave the data where it is, where it lives, where it thrives and bring AI to the data. And so what we are putting into the next generation of CloudPak for data is an intelligent data fabric, right? That is fueled by AI and that is going to abstract a lot of the complexity out of all of this. Let you keep the data where it's at and be able to discover that data intelligently, be able to catalog it, be able to understand it, right? And more importantly, be able to do unified queries and updates across all these distributed sources of data and bring the records together without having to take weeks and months to build new data pipelines. And across that entire ecosystem, be able to enforce universal privacy and usage policies, which is absolutely critical. Forced estimates at 50% of the data is not used because they're afraid that it's going to break policy. Oh yeah, I mean, that's a huge trust issue. I mean, I was talking to a practitioner and he's like, you know, we don't even want to do some of these transactions that are interesting experiments and cloud opportunities because of the compliance risk. They're afraid to get sued. Yeah, no, that's right. And each one of those data stores, just think about the ecosystem we're talking about here of sources and consumers, data consumers, AI consumers, and of course, all the sources that are siloed all over the place. A lot of these repositories and a lot of these different cloud environments have different policies in terms of usage and privacy, right? So how do you bring out all that together? And what we're delivering in the next version of Cloudpack for data is a universal privacy plane, if you will, which is called auto privacy. And it will basically abstract all the complexity of the different policies, allow you to create them and enforce it universally. And you couldn't imagine the productivity of being able to deliver that versus having to hand deal with this in a manual way. That's an example of what a data fabric would do. You know, what's interesting is you're getting at these, I mean, I'm hearing the conversation about the solution. It's okay, I'm not in mind going, okay, what's the benefits I hear? I hear speed, I hear ease of use, compliance, trust, but what you're really getting at is agility. And there's a upside for agility that's moving fast and taking advantage of new opportunities for automating something away. But you mentioned the trust piece because that's where I see people afraid, like, okay, if I move too fast, will I trip on over or some governance issue? Like that's a huge thing. This is a big problem. It's a massive problem. I mean, I think there's four, four areas from a business perspective, right? One is think about digital experiences. And we know that six in 10 customers that defect from a brand because of some bad experience usually don't return. And it's estimated that's costing the industry, you know, close to $500 billion a year. So optimizing that digital experience, which is all driven through personalized experiences, responsive experiences, which is you have to bring the data together to be able to do that, right? The second is the regulatory and reputational risk. That's another 180 billion or so, which in many cases is 8% of revenue, just to mitigate all that risk of using data, not only regulatory but reputational. Just think about lost productivity. How many hours every week is a worker doing mundane tasks, low value work, because it's not automated. That's like another hundred or so billion dollars of cost for enterprises. You can go on with inaccurate planning and forecasting, supply chains being inefficient. All this is being fueled by the data, right? So the more you can bring all this data together, unify it, create new views that are aggregate in nature and uncover hidden insights that you couldn't do before, that's the magic sauce here, right? Well, my last question for you on this product before we wrap up is there's a huge trend towards ecosystem network effect integration, right? There's more integrations, people are partnering. I mean, you have solutions where that rely on different people in the supply chain or value chain of a solution, whether you're a concession at a ballpark or an enterprise, you're connecting with other APIs. This is cloud, right? How does your cloud pack for data handle that integration and that trust because this is really the deployment scenario, your thoughts? Yeah, I mean, I think the core of cloud pack for data is it's going to greatly enhance productivity. It's going to lower costs of these, you know, complex data states, it's going to lower risk of all this and it's going to help you uncover hidden insights that you couldn't see before, not only because of AI, but because we unify the data to get more out of it. We then go on to really point out that it's a truly open platform with an open ecosystem. So we are partnering with all the cloud partners, right? We have a vast network of software providers that can extend and endlessly customize the platform. We have integrator partners and it's all based on open source communities. So it is fully extensible and it's customizable to unique needs of every customer on any cloud you want or across any cloud you want. All right, Scott, that's great stuff. Thanks for coming on theCUBE. Great to see you. Scott Hapner, Vice President of Marketing at IBM for data and AI, the hottest area. Great, great CUBE alumni, great insight. Thanks, Scott, for coming on. All right, thank you. Okay, I'm John Furrier with theCUBE. You're watching IBM Think 2021 coverage. Thanks for watching.