 From Times Square, in the heart of New York City, it's theCUBE, covering Imagine 2018. Brought to you by Automation Anywhere. Welcome back everybody, Jeff Frick here with theCUBE. We're Automation Anywhere in Midtown Manhattan, 2018. Excited to have our next guest, he's Kevin Crow and he's partner of Financial Services Intelligent Automation Leader at PWC. Kevin, great to see you. Thank you. So Financial Services seems to be a theme, we're here in Manhattan. Why is Financial Services an early adopter or maybe a frequent adopter or an advanced adopter of the RPA technology? Sure, so I think as we see our Financial Services clients and their agendas, there's been a huge focus on productivity and simplifying their overall operating model over the past couple of years. Banks in particular have gone through several years of having to focus a lot of their spending on non-discretionary manners like regulatory compliance and risk management. And what that's generated is a need as they start looking towards the next generation to really start thinking about what they're gonna look like in a post-regulatory environment and automation is quickly risen to the top of that. That's gonna bring in a post-regulatory environment. Yes. Why a post-regulatory environment? Well, I mean, if you look through what banks have had to deal with in terms of Dodd-Frank, in terms of C-CAR and other regulation from the Federal Reserve, these are things that took a lot of spending both on implementing operational processes and on implementing technology. A lot of that work is starting to, the banks are putting that behind themselves. And so what, as they look forward and look at how they're going to gain more profitability in the future, the challenge becomes there's not necessarily a new set of product innovation coming in. And so you have to really look at the expense line. And so because of that, automation has risen to the top of that agenda. And so this continues to be one of the top areas of interest that we're getting from our clients. So in each day post-regulatory, you mean like a new regulation that they have to respond to, not that there's suddenly not gonna be a regulation. There's not a lot of new regulations coming in right now. Especially. I said that pesky one last week, GDRP. Yeah, but in the U.S., we're in an environment right now. I mean, we just, you know, there was just, you know, the revisions to the Dodd-Frank bill that were passed a lot of regulatory rules are actually being loosened. So, you don't necessarily have an increase in dollars that are going to be going into that. Right, right. So it just always fascinates me, right? I thought ERP was supposed to ring out all the efficiency in our systems. But that was not the case, not even by a long shot. And now we continue to find these new avenues for more efficiency. And clearly this is a big one that we stumbled upon. Yeah, you know, I think it's interesting when you look at kind of big technology investment over the last decade or two, you could argue a lot of effort's been focused at what I call the kind of core infrastructure and core plumbing. So, you know, how do I consolidate data into a single location? How do I make sure that data records house the different parts of my organization? But that like kind of last mile of what someone does as part of their day-to-day business process was never really addressed, you know, or was only addressed in pieces. And so I think as you start looking at the productivity term and how you actually start getting efficiency, we have very few clients that are saying, I want to take on that next big ERP type implementation, you know, or I'm ready to spend $300 million on a new project. They're looking to try to get the most value out of what they already have and they're actually looking to look at that last mile and how can they actually get some benefit off of it. So, you know, RPA technologies, I think we're one of the catalysts of just being the perfect technology in the right place at the right time from a current business environment and current technology spend perspective. Yeah, it's pretty interesting. We here was talking about, you know, one of the big benefits is that you can take advantage of your existing infrastructure. You know, it's not a big giant ribbon-replace project, but it's again, it's this marginal and incremental automation that you just get little benefit, little benefit, little benefit into the day turns into a big benefit. Yeah, and I think that's, you know, it's quick, it's fast, it's, you know, it could be implemented in an agile manner and, you know, our clients are continuously telling us over and over again, they're willing to invest, but they want to invest where they're going to see a tangible payback immediately. But I think when you start to talk the concept of digital transformation, it can mean a lot of different things to a lot of different people, but there are big picture changes that could be made, those that may be longer term trends, but there are more immediate things and more immediate benefits that could be gained. And I think that's really the sweet spot of where RPA and Automation Anywhere fall into. I love, I was just looking up, Jeff N. Milton's keynote said, this is the easiest found money of any digital transformation project. I think that was the quote that you'll ever do. That's a pretty nice endorsement. Yeah, and it's, you know, as we go out, we talk to CFO, COO, CIOs, you know, it's, the value proposition is really attractive because, you know, there's a track record of failed, technology projects failed, big transformation projects, and no one wants to necessarily risk their career on creating the next big failure. And so I think, you know, using technology like RPA almost as an entry point or kind of a gateway drug into the digital world, see the benefits, you know, start to understand what are some of the business problems and historical kind of, you know, things you're trying to untangle in your infrastructure, attack that, and then, you know, start to layer on additional things on top of that once you get good with RPA. And then you can start figuring out, okay, that's the gateway into artificial intelligence. Okay, now how do I start to apply AI across my organization as you get beyond AI? Okay, how do I get into, you know, more advanced data infrastructure? And you can start thinking about this world where you're just going to, you know, rather than do the big five-year project or you're going to try to solve world hunger, it gives you a chance to kind of incrementally go digital over time. And I think that's definitely the direction we see a lot of our clients wanting to go in. Kevin, I want to get your feedback on another topic that came up again in the keynote was just security. You know, it was like the last thing that was mentioned, you know, ABC, the EFG and security. Financial services obviously security is number one. It's baked into everything that everyone's trying to do now, it's no longer this big moat and wall. It's got to be everywhere. So I'm just curious from the customer adoption point of view, where's security come up in the conversation? Has it been a big deal? Is it just assumed? Is there a lot of good stuff that you can demonstrate to clients? How does security fit within this whole RPA world? You know, it's security and I would just say the broader kind of risk management pieces to RPA infrastructure are one of the first questions that get asked in a highly regulated environment like financial services. You know, the technology is easy and powerful with RPA, but you also have to take a step back and say okay, I can program a bot to go do anything in my infrastructure and that could mean running a reconciliation or it could mean going to our wire system and trying to send money out the door. So there's a lot of concern around not only understanding the technical aspects to, you know, how the tools work with different types of security technologies, but more looking at your approach to entitlements and your approach to how you actually manage who has access to code bots, to put bots into production, to, you know, over time understand what happens. You know, we did a presentation to a board of directors a couple months ago on kind of automation more broadly and, you know, this is, you know, senior level executives. The first question we got was, you know, okay, how do I prevent the 22 year old kid that just came off of campus from building a bot that no one knows about, setting it loose in our infrastructure and it going rogue, right? And so, I mean, and this group was pretty sad, they caught on to it very quickly and, you know, the CIO of this client was sitting next to me and she kind of, it didn't have an immediate answer to that and so I think that was kind of the aha moment. This is something we really need to put some thought into around, you know, who are we gonna let build bots? What policies are going to be set around how bots get deployed into our production environment? How are we going to monitor what happens? You know, how are we going to get our auditors, our operational risk folks, our regulators? How are we going to get all of our different stakeholder groups comfortable that we have a well-controlled, well-functioning bot, you know, bot infrastructure that exists? Right, because the bots actually act like people, they're entitled as like a role, right, within your organization. We have clients that have literally had to set bots up as new employees, like they get onboarded, they have a, you go to the corporate directory and you can see a picture of R2D2, right, like and it's the way they get around, you know, how they get a bot intelligence system but it's still, it's not a human, right, so you still have to have a policy for how you actually will get code that uses that bot entitlement to function, right, so that has to be done in a well-disciplined, well-controlled manner. Right, because to give them the ability to provide information to help a person make a decision is very different than basically enabling them to make that decision and take proactive action. Exactly. Yeah, it's funny, we talked to Dr. Robert Gates at a show a little while ago and he said the only place in the US military where a machine can actually shoot a gun is in the Korean border, but every place else, you know, they can make suggestions but ultimately it's got to be a person that makes a decision to push the button. And we're seeing, you know, trying to equate that to financial services, you see a similar pattern where there are certain areas where people are very comfortable playing this technology, you know, you get into, you know, accounting and reporting and more back office type processes. You get other areas where people are a little bit less comfortable, anything that touches kind of wire systems or touches things that, you know, going out the door, touches the concor trading processes, things like that, you know, there's a different risk profile associated with it. Right. I think the other challenge too is, you know, RPAs is getting the gateway drug into this. You know, going back to my previous point, as you start to layer additional technologies into this, you might have less transparency over understanding clearly what's happening, especially as artificial intelligence takes a much broader role in this. And so there's going to be a lot of scrutiny, I think, over the next couple of years really put into like, how do I understand the models that are created, you know, by artificial intelligence technologies in those decisions that are being made? Because you, you know, if your regulator says, okay, why did you make this decision? You have to be able to explain it as the supervisor of that intelligent bot. Right. You can't just say, oh, it's because it's what the machine told me to do. Right, right. So that's, yeah, it'll be one of the interesting challenges that's ahead of us. Yeah, it's good. I mean, it's part of the whole scale conversation, that interesting conversation the other day where they got, you know, talking about really opening up those AI boxes, you know, so that you have an auditable process, right? You can actually point to why it made the decision, even if you're not the one that made it in real time and is doing it really, really quickly. So, really important piece. Yeah, and as PWC, it's one of our challenges. As a consultant, I'm helping clients implement this. My, you know, my colleagues in our audit practice are now grappling with that same question because we're increasingly being asked to audit that type of infrastructure and have to, you know, prove that something did what it was supposed to have done. Right, right. All right, Kevin, well, nothing but opportunities for you ahead. Yes. Thanks for taking a few minutes to stop by. Okay, thank you for having me. All right, he's Kevin. I'm Jeff. Thank you for Automation Anywhere Imagine 2018 in Manhattan. Thanks for watching.