Podcast transcript: How to drive meaningful business impact with intelligent automation
40 mins 35 secs | 6 December 2023
Susannah Streeter
Hello and welcome to the EY and Microsoft Tech Directions podcast. I'm your host Susannah Streeter and in this episode, we're going to lift the lid on how to drive value through increased automation and harness the power of AI. The race is on for companies to automate their processes to increase efficiency and productivity. The aim is to reduce the need for manual labor and human errors. Automation and advanced machine learning should allow companies to scale operations, streamline workflows and allocate resources more effectively. The prizes should be significant, cost savings and competitive advantages in the market. But it will often rely on getting the right tech and organization in place to optimize the opportunities which exist. So, in this podcast, we're going to delve into the best ways of doing this. We'll explore prebuilt integrations, out-of-the-box AI capabilities, and implementing generative AI strategies. These are big topics, to say the least. So, I am delighted to tell you that I have three experts in the field waiting in the wings with many years of experience between them to provide you with some really valuable insights. But before I introduce them, please remember conversations during this podcast should not be relied on as accounting, legal, investment, or other professional advice; listeners must, of course, consult their own advisers.
So now I'm really pleased to introduce William Smith, Senior Manager at EY LLP, specializing in AI-enabled automation, advanced analytics and digital enablement. Where are you joining us from, Billy?
Billy Smith
Hey Susannah, I'm calling from New York. Great to be with you.
Streeter
Lovely to have you with me. And also let's welcome Jack Virdee, who's VP of Finance and Automation at Omnicom. Jack, where are you today?
Jack Virdee
Like Billy, I'm just outside of New York in a bit of a soggy situation here today.
Streeter
It's been a bit soggy here too, as well today in fact. I'm talking to you from Bristol in the UK. But let's welcome Stephen Siciliano, who's VP of Power Automate at Microsoft. Stephen, what's the weather like where you are?
Stephen Siciliano
Well, Seattle is actually nice and sunny today, so no rain in our forecast.
Streeter
Good to hear. Stephen, first of all, tell me what you see as the most striking trend in the automation market right now.
Siciliano
Well, this hopefully won't come as a huge surprise to folks. But I think it is the way that AI is really transforming the way that all business processes are going to run in the near future in organizations. So what has happened over the past couple of years is we've come to realize how large-scale foundational AI models that are generative, can really start applying to processes in unique ways. You know, before in the automation space, everything had to be very clear, setup up front rules-based. But now it's possible for AI to actually solve core business problems in ways that would have required human thinking and human interaction before. So with this transformation, I think that the number of scenarios that can be automated is going to just explode way more than ever existed previously. And this is not just, of course, about classic business processes. This will also enable net new things that are just day-to-day processes that people work with every single day to be automated because people are going to have these AI bots that can help them build and run their automation in ways that were impossible before.
Streeter
So, lighting up whole new ways of working. Well, let me bring in Billy. What are clients telling you about their motivation right now about jumping into this trend?
Smith
What clients are telling us is that it's all about their competitive differentiation in the market. So when we're talking to our clients, what we're seeing is that they're constantly looking for those ways to extract as much value as possible from their process, people and tech. And one of the ways that we're proving, and companies see it, is that automation in digital is where a lot of that differentiation, when it's wielded the right way, can really occur. And you know, it's becoming really apparent. So when we go into a client, we're assessing where they're at, and then we build out those bots and products that really transform the way that they do things. Whether it's through enhanced processing or better UX or better data visualisations, what we're seeing is that companies are constantly looking for those impactful builds. And that that really gets them to start the journey with us and jump right into automation.
Streeter
To what extent is it a first-mover advantage?
Smith
There's no doubt about it, that there's a lot of first-mover advantage, but the one thing that I'll say is that, you know, it's important to move fast, but at the same time, it's equally important to set up the right governance and guardrails to be able to scale. 100% yes, clients should really be looking to move fast in this area and get that advantage, but with that caveat to say, we need to be very strategic and pick the right use cases and capture that value.
Streeter
Jack, why do you think some companies are still hesitant about taking the plunge right now and immersing in automation?
Virdee
I'd be very surprised if some of the companies that are out there aren't already dabbling with automation in some capacity. However, I think the trend that we're starting to see now is that large organizations like our own, are starting to finally join these disparate offerings and services together in a more meaningful enterprise way. I believe if you're still hesitant, you're probably waiting to see the landscape, hoping it will settle down. I think the news that I have, and it kind of echoes what Stephen was saying, is that the number of scenarios that are out there that come to the forefront with generative AI are a reason why you should actually jump in. I agree with Billy wholeheartedly that what you ought to do is lay the foundation so that you can have the first-mover advantage. But I also believe that with the platforms and the right partners, you can do both. You can actually take advantage of the first mover advantage pretty quickly now versus waiting in line and waiting for things to settle down because you're going to miss the boat.
Streeter
So it's never really going to settle down, it seems. So why did you move when you did what was your business need? What prompted you?
Virdee
The reason we moved really quickly goes back to 2020. COVID-19 pandemic, as probably a similar catalyst for many businesses, was a major catalyst for our business and a lot of our digital transformations. And in this case, automation and, and process engineering, really were the levers that we wanted to press upon. So, what we needed to do was insulate our business from human capital dependencies and get serious about automation. Therefore, we activated at an enterprise level, our program needs to do automation. For us, the value of automation is not limited to cost savings alone. But we were also able to unlock insights that were previously trapped below the surface.
Streeter
Really fascinating. Billy, tell me about the different ways that the EY teams can work with a client in automation.
Smith
So we provide a variety of different types of projects and services for clients, you know, based on where they are, you know, starting from scratch, or if they're even mature in the capability. And those services, they're all the way from strategy engagements to system implementation type services. So, just to give a little bit of a feel for each one of them. The first type of project is setting up a COE (center of excellence), or an operating model for our clients. So essentially, going from zero to one is what we do for a lot of clients just starting out on their journey. We'll set up an operating model, intake mechanisms, rules, and responsibilities. The next area is really around the implementation for a client. So here's where we'll work to stand up dev pods and sprint teams and resource them with teams of BAs, and developers, PMS, architects, really everybody that needs to be involved in the implementation and actually putting in the tag. One thing that the EY team does a really good job - is that we embed sector and domain and different process SMEs within those dev teams to make sure that we're getting to the core of the problem, really help solving and automating the right type of solution. And then the last one I want to highlight is really more strategy-aligned type projects. And it can take different forms, like demand generation assessments, or value case assessments. I've even done some work for due diligence for buy side programmes for PE. And that's really exciting work. Those are the main types of services. And oftentimes, we work with clients in all different realms. What we did with Jack is we helped establish their CO, we did some opportunity assessment and found these high impact use cases. And then we also implemented using Power Platform. That's just a feel for some of the ways that we help our clients.
Streeter
Well, let me bring you Stephen, how do you see the market right now for Microsoft and particularly Power Platform?
Siciliano
Great question. I think the market right now is, in some ways, growing faster than it ever has before. What we've seen in the past is there has, of course been automation for many, many, many years, right? The investments that companies have been making and technologies like RPA, like digital process automation, have been ongoing and a part of most CIOs' portfolios for quite some time. So, automation itself isn't particularly new. But what is new is the breadth of scenarios that can be enabled with automation. I think that's something that we at Microsoft have been really focused on with Power Automate, and the broader Power Platform is making automation accessible with a low code approach. And the low code approach is so important because it says, Hey, if you are a business user, if you're somebody who is most familiar with that business problem, you can actually solve that business problem yourself, you can solve it using tools at your own disposal, without having to go to central IT and getting put into a backlog that takes six months to be reviewed and prioritized, you can just do it yourself. So that is where the market absolutely is going across the board. Not just Microsoft looking at this. But this is something that I think everybody in the automation space is very passionate about. And then of course, with Microsoft, what we've done is we've, for example, created the copilot that sits alongside people trying to build out their automation so that they can build automation more easily. They don't need to be as deep of an expert, and they can roll out that automation, and make sure that it actually works and is resilient because the copilot can help enforce those best practices and understand what it is that people are building. So I'd say that's really the main trend that we're seeing is this broadening of automation. It's not that automation is new; it's that it's now more accessible than ever before.
Streeter
Absolutely, I'm so looking forward to having my personal copilot. But Billy, what kinds of trends are you seeing when it comes to the gravitational pull of certain platforms?
Smith
I'll name a few of them that come to mind that are starting to really take the fore. So, the first trend I'm seeing is that a lot of the successful platforms are starting to offer a full suite or ecosystem of digital capabilities within their umbrella. They're not so single-threaded in terms of what they offer as a main product. So when it comes to Microsoft and Power Platform, and why I believe they're getting a lot of momentum in the market right now, out-of-the-box clients, they're able to get power Automate desktop and cloud flows for process automation. You get AI builder for document intelligence process advisor and Power Apps for more front-end development. The second trend I'm really seeing is that platforms that are starting to invest in building out those prebuilt connectors or integrations are really winning in the market. So inherently, automation, COEs were dot connectors, right? So we're interacting through front-end screens or interfaces, and we work with all the system landscape within different environments. And when there are those prebuilt APIs or connectors, it really dramatically improves the stability and speed to value for implementing bots. Clients are starting to see that, and they're looking for, how can I get to speed as fast as possible. And how are they going to be as stable as possible? And the last one - a lot of clients are looking to minimize their ongoing costs with the opposite side of implementation here. And the platforms that don't solely rely on for example, virtual machine architectures, to scale up is a real differentiator. Microsoft are starting to invest in cloud flows versus Power Automate desktop, and that's really paying dividends for them, and helping minimize some of that ongoing costs that clients start to see when they implement this technology.
Streeter
And, Jack, tell me about how you set off on your automation journey. And why did you make the choices you did?
Virdee
Similar to what Stephen and Billy were saying, I think, the foundation that we focused on was two areas: the platform in the process, and we realized that the structural components would hopefully follow as we build out the former. And that's where we are right now. First, on the platform side, we needed to align on the unified platform, as I was saying, we had disparate and unjoined offerings across our network of 1000 agencies. We needed a unified platform to help us tie that all together. We put Microsoft into the pitch driven by their growth in the RPA space, and a major reliance on their infrastructure within our organization. We're a very heavy Microsoft shop. During the pitch, what we ended up doing was finally choosing the Microsoft platform due to the growth in the end-to-end spectrum that Billy mentioned. So being able to, as Stephen mentioned, the low code, no code opportunities, and then coupling that with the machine learning and then eventually coming into the Azure Open AI offerings that more recently have come to bear, is a reason that I think we actually got quite lucky with the Microsoft platform. So the Microsoft platform and the partnership that we have with Microsoft, we've been fortunate that it's enabled us to create some first-mover advantages that we're pretty thankful for. The second part of our automation strategy - we're setting up a Centre of Excellence with EY, setting up the governance, coming in with the process documentation, helping us define the structure. We bought the full cake from EY. And what we also appreciated from EY was being able to benchmark this against what other clients are doing in this space. So we did three things. One, we had EY help us build the entire Centre of Excellence for us. Two, we had EY, using the Microsoft platform, help to create an orchestration application so that as any use cases came through, we kind of said, how do we eat our own dog food to some extent here? So what we wanted to do was any use case that came through it would go through a Power Platform-built orchestration app, and that is in production right now. Then we also took EY and their consultancy, to help us define 10 use cases out of 100, that we had thought of together in a brainstorming session, and apply that using the Power Platform, and then develop those and put those out there to use. So those 10 have now grown to 10x. So we have about 100 in the queue. And that's within six months, I think that has to do both with the foundational elements that EY helped us create from a governance model, as well as the extensibility of the platform. The fact that we can create applications and solutions for our agency in such rapid fashion speaks volumes for both the process as well as the platform that we built together.
Streeter
Yeah, absolutely. The sheer number of potential use cases you've built up already is quite remarkable. And Billy, you talk to clients all the time about the value of automation, how do you articulate it? And what metrics do you highlight?
Smith
Yeah, 100%. And we get this all the time. And before we get into the specific metrics, the number one thing and reason why a programme may not be succeeding or generating that value is really because of two reasons. One, they may not be selecting the right use cases. Or second, they're not articulating the full value that automation really provides them and getting to that right metric. So a couple of the things that we do in this area, and we worked with Jack and his team and making sure that we're going by these principles. The first thing that we do with companies is we look to define the specific drivers and articulate the full value automation provides. A common mistake here is that companies sometimes only value that direct ROI and the time savings back to the business. And when you think about it, that's only really a portion of the value that automation provides. We will work with our clients and account for factors like speed and velocity or throughput advantage or de-risking and what that means to the value of a specific use case. And then be able to articulate that, as well as the time savings back when we implemented use case. The second thing is what we do is we start with a KPI. We don't do it midway or after an implementation has taken place; we'll actually take that value and approach right up front. It's referred to as our micro transformation type methodology. What we found is that when we focus on value and strategic metrics, it dramatically shifts the impact on what we're ultimately able to deliver for our clients and build the most value for what they're trying to accomplish.
Streeter
So you're really scaling a mountain of value and really pointing out a value that some of your clients simply didn't know could exist. Jack, do you feel that your journey was successful because you did highlight the right metrics at the start? Or were certain metrics pointed out to you that you just hadn't thought about?
Virdee
I think it's definitely the latter, Susannah. I think we were fortunate to have EY as a partner in the beginning because I think there's a continuum of the metrics story. So first is setting up the right metric in the beginning of a large initiative like this, which is what are the metrics you need in order to gain executive sponsorship, and those really revolve around savings, ROI, timelines, etc. At a program level, EY was able to help us rationalize and put that in front of our senior management enough that it created the runway for us to get going. Then, when we get into the individual use cases, as Billy had mentioned, and Susannah as you highlight, the EY team helped us absolutely to uncover these other what EY calls colors of money that we didn't think about. When we typically think about value, we think about cost savings or hours back. That dynamic has shifted from both hours back to improvements of accuracy, improvements around decision making, other values of intelligence that was otherwise unlocked or hidden from us, that we were able to unlock. The third aspect, and this is just where we are in our journey, and this is the more difficult one, is now that we've defined the metrics on the front end of the use case. It's actually figuring out what the metrics are, and hopefully, they're the same, when you've delivered the automation, and you put it into production, and finding potentially an appetite from the business to quantify and actually put that value in hand. So if it is improved accuracy, what is the mechanism by which we're actually going to measure that the accuracy is better, as opposed to just putting automation out there and saying, this should have made your life better? Equally, you need to have those metrics. And we relied on the EY governance model to say, every 30, 60, 90 days after an automation is live, there was a framework that they walked us through so that we could start to measure those metrics, even on the back end.
Streeter
So, you really do need to choose the right partnership to gain the most out of this process. And Stephen, to the extent you can share, where is the platform going now in the years ahead with the product roadmap? This will be a real interest for anybody starting out this journey or perhaps already along the route.
Siciliano
I think there's a lot of exciting things that we have on the docket. But I'm going to bring it back to the one big area that we've been focusing on recently, which is AI, right? With the innovations that have been happening with these large language models, generative AI, we have a whole new set of things that we can really automate, unlike what was possible before. And I think we talked a little bit about understanding the right metrics and understanding how to measure them. I think that in many cases, though, people don't even understand what to automate in the first place. And AI will be able to look at the data that you have inside of your organization, look at the existing processes that are there, look at the tasks that people are doing, and be able to make proactive recommendations on how things can get better, right? How things can be automated in the first place. And that's ultimately part of the core value prop of automation is to remove those bottlenecks that exist inside of processes. And these could be big processes, small processes, processes of any scale. According to some stats, half of all digital work today can be automated, which means that there's this massive opportunity, but people just don't know that they're even capable of automating in the first place. So, in the future, we are going to work collectively with all different parts of the Power Platform to bring together more intelligent AI copilots. It's not just about strictly power automate, and cloud flows. But it's also about bringing in power virtual agents. It's also about Power Apps, it's about leveraging power pages, all of the different parts of the platform to solve these needs, and copilot will be able to span and bring together solutions that touch on each of those areas. So these holistic solutions - sometimes we hear the term hyper-automation. This idea of really focusing on the business problems and using all the tools in the toolbox, that's going to be a way that AI is going to help to solve all of these needs. And ultimately, what we see is something like 80% of all processes that are going to be built out are going to happen outside of central IT as this technology trend continues. And that's because there will be recommendations, these AI capabilities, and ultimately, the tools like the Power Platform that are accessible for the citizen developer to go and build all these things out.
Streeter
Well, given this 80% prediction that Stephen's been talking about, Billy, how should businesses evolve to deal with that change?
Smith
Yes, great question. And that is a good point by Stephen. To me, it's about businesses building out their automation operations, so that they can be as flexible as possible to work with all different types of groups and users within an organization. Typically starting out, companies build up their operations and their COEs from the center, right? That's where a COE typically sits right within central IT. But as companies evolve, and groups get more and more mature, and they start to build out their skill sets, that central capability has to be flexible so that they're able to cater to all different types of groups and establish their own capabilities. So once again, I'd say it's important that the COE sets rules that aren't so rigid and allows for that federated or citizen development capability. And it's just becoming more and more important, especially as copilot starts to get disseminated across organizations. And Power Platform, which is no code low code, is starting to enable a lot of that citizen development.
Streeter
So Jack, where do you see Omnicom going with automation within the next one to two years, and how are you driving that at the center of the organization but also right across the business?
Virdee
So we're seeing AI serve as the tip of the spear for unlocking a lot more scenarios across the organization, and the appetite to actually solve business problems is preceding what would typically have been a platform rollout to simply just have this as a tool that other organizations can give to their IT group to put in front of their, their agencies and hope that they use it. We're embracing this innovative thinking and converting it into a first-mover advantage as much as we can. All our current use cases rely on the full stack of the solution that Stephen has mentioned. So its Power Platform, including all facets of that. So it's Power Automate, Power Apps, it's the virtual agents, yes, it's going to evolve and get a lot better with copilot. But the power of the stack is actually in the interoperability between not only just the power ecosystem, but also further into the Azure echo system, and the API calls that it can make across other legacy platforms that we have within our organization. So we're not beholden to only one platform. But we're finally starting to see the light at the end of the tunnel where systems can start to talk to each other. And now you add on top of that the generative AI ability, and we start to drive intelligence from one platform to another, that simply was always hidden away. As far as how we're going to evolve this. As I mentioned before, we've created a center of excellence across the enterprise. As Billy mentioned, we predict that this will start to move into a more distributed model. However, I think we're probably going to come back a little bit short of a full citizen developer. I think what we want to probably do is create scale across our business units, and still have some level of governance model within those business units rather than an individual creating a single automation for themselves. What we would rather do is start to create automation that can then scale across the 80,000 people that we have.
Streeter
And Billy, looking ahead. Do you see the business then shifting when it comes to high-value use cases?
Smith
It's a great question. It's so important to find those high-value use cases. A couple of things I'll say on that. I think that automation departments and CEOs, they're going to have to keep on doubling down in terms of partnering with the business and finding those strategic builds, that are able to be those impactful high-value use cases. Working with a lot of clients, they're exponentially more important to find those real, impactful things that change the way of working. And then the second thing from a tech perspective, that ecosystem of digital capabilities and finding high-value use cases, what we're seeing is that the value of good architecture is really huge. To be able to architect the products in a way that you're able to extract as much value as possible while using all that's available to you, within all of these digital capabilities, that's going enable you to find those high-value impactful builds. So once again, you're going to start partnering with the business and also really putting an emphasis on that architecture to get to the high-value use case.
Streeter
Stephen, let me bring you back in. What are you most excited about in terms of AI and how Microsoft is driving the capabilities?
Siciliano
I think the biggest thing for me is the accessibility of these capabilities. So you know, really, the idea is that if you take AI, for example, you don't have to be a PhD with 10 years of experience building out complex AI models, in order to be able to use it as a building block in your workflows. Some of the things that we've seen people doing with GPT inside of Power Automate, for example, have really blown my mind. For example, a customer support engineer can create a workflow that can take an incoming customer inquiry, process that inquiry, and auto-generate a fairly good response by just passing the context of that customer and the incoming query to Chat GPT, have chat GPT generate a well-formed appropriate response. That can save that customer support person, 5, 10, or 15 minutes of looking up data or typing text. Now they can just have five seconds of reviewing what the GPT created and hitting send, and they have just made themselves way more productive. You didn't even have to really know what GPT was in the first place, it's just something that they can ignore. All they have to know is they can have this action in their workflow that can do all of that intelligence for them. Frankly, we can't even imagine all the different types of scenarios that people can use AI for. This is why we've made these tools so accessible. And this is just built in. They don't have to go sign up for an Azure subscription. They don't have to go learn all these other ways of doing things. They just use AI builder right inside of the platform and can build something. We're really excited about how, for the first time, this is something that's at the fingertips of millions of people.
Streeter
And I suppose you must be really excited about seeing the creations in the future that come out of this. Jack, what are your thoughts about generative AI and just how it can integrate with your existing plans?
Virdee
Absolutely. And I would echo Stephen's enthusiasm here. We are equally very enthusiastic and stepping into this as much as possible. I think what Stephen had touched on as far as reducing the barrier to entry here is very real. It has made automation a lot more palatable within the organization, something that our employees can finally wrap their heads around without getting involved in the coding elements that used to be the case with a lot of the previous platforms. And I know that they're all maturing towards this area, but generative AI finally puts it into what I'll say is common speak. And that's a huge barrier to overcome within an organization. So what we also see as a benefit from generative AI, I mentioned it earlier, is this notion of enterprise intelligence. That starts to become a huge value within 80,000 employees that all ideally are working the same way. But the truth be told, they're not. They're in different markets. They're using different languages. And we can finally find a way to actually unlock and democratize the knowledge that's trapped within these agencies. One example is we have multiple markets where they're doing work, but because of language barriers, we never knew what work they were doing. So we've started to use generative AI to create summaries of that knowledge base and actually start to learn and unlock the value of work that was done in Brazil or Japan that you never knew about. Because we only focused on the English-speaking content that we used as kind of our knowledge base and best practices. Now we can really start to operate as a global company.
Streeter
Really fascinating how knowledge bases can be expanded so rapidly. Billy, what's your take on this? Do you think that companies are geared up to provide the rescaling needed? This is, after all, going to necessitate brand-new ways of working.
Smith
The answer is definitely yes. They're gearing up and putting on the gas to start a lot of that retraining and skilling up their resources. The platforms are really helping out in terms of the user experience and an easy product to pick up, as a nascent user of some of these platforms. So for copilot, I think that's going to be disseminated across the business very soon, and helping a lot with the reskilling, especially when creating your own code, and helping disseminate that. And with Power Platform, I know we touched on it a bit earlier. But the fact that it's low code, no code, there's a lot of courses enabled - you can do self-certifications. It's really helping with that more citizen development mindset.
Streeter
Jack, I'd just like to ask you, though, how are you navigating the fact that there is still a lot of fear around about where jobs will end up with AI? And just how many might go?
Virdee
There's obviously a lot of hype out there. And I think that some of that has to do with any time that you go through a digital or an operational transformation. But, I do believe that solid change management and an empathetic change management program are necessary for anything like this. So the reality that we're finding is that along with efficiency, the good byproduct is that we're finding a lot more time to do the work that we really wanted to do. So the way that we broach this is it really is more complimentary rather than replacement. Gone are the days of being bogged down in spreadsheet creation or initial coding, and we can now focus on the editing and the more difficult business challenges. So we all are working across multiple facets of our 80,000 team to help them from an education perspective, all the way from the top of the organization, leveraging, as Billy had mentioned, some of the existing platform material and making sure that that's mandatory for all of our employees. But also creating two facets as well, making sure that there's a strong communication channel. And then also increasing the transparency of what's going on. I feel like what ends up happening is people build into the fear side of this change when they don't know what's going on. If we train, communicate, and create an environment of collaboration and openness, then we end up with a good product on the other side, where the employees are actually helping to drive those scenarios rather than the change happening to them. They're really doing it with us.
Streeter
So Stephen, it is really hard to cover this huge subject in just one podcast. What do you recommend to listeners who want to learn more about scaling up automation?
Siciliano
So there's a tonne of really exciting content online. So to start with, of course, if you go to the Power Platform websites and the Power Platform community. We actually have millions of people in the community who are sharing their ideas, sharing their experiences. We also have a number of really, really great partners. Who have spent years and years developing deep expertise in the best practices and the best ways to run automation, set up that Centre of Excellence, like we were talking about, and putting those guardrails in place that make sure that the team members feel confident. I'd say working closely with the partner community is a really great way for organizations to upskill and learn. And there's also a lot of public content out there. So I'm always impressed with, for example, how much there is on YouTube, of people just sharing their ideas, doing walkthroughs, and talking about their strategies for automation in the organization. So even those types of kind of public forums, I think, are a great place to go to learn and get upskilled on the opportunity here.
Streeter
Okay, well, thank you, Stephen. And we are nearly out of time. But I just want to get one final thought from all of you. Can you peer into the future and tell me just how you see automation developing over the next decade? Can you see the horizon ahead, or is it still too difficult to assess, given just how fast the tech is developing? Billy?
Smith
It's all about AI. And we're already seeing it. It's the extension of what's already been built over the last 2, 3, 5 years, and it's starting to enable more natural language processing (NLP), really transforming different domains while using the new tech. I'm so excited to see what the future holds for every department that we work with now that they're really rethinking the way that they do things with the emergence of AI.
Streeter
Jack?
Virdee
I would absolutely echo what Billy just said, I think it is about AI and automation continuing to be the tip of that spear. We're pretty bullish on where I see this going. I think broadly, the notion of automation will eventually come to very much like we think of the internet. And it's simply integrated into the ethos of how we live our lives, we don't think of the internet in a certain way. And going forward, I think we're going to stop thinking about automation. And just think about it as, here's a problem; how do we solve it? My advice to the group would be to set up your platform and your Centre of Excellence now because the future is going to build upon those, and it's pretty exciting.
Streeter
And finally, Stephen?
Siciliano
From my perspective, if I had asked myself two years ago - would we have anything like what we have today, I couldn't possibly have predicted what we've gotten to with all these new AI capabilities. So at the risk of repeating what we just said, I would say AI is the biggest trend, but I don't think we know yet where this is going to go. I think in two years, in five years, in 10 years, we are going to be constantly surprised with the ways that automation can transform people's lives. And I do think it's going to get to a place where automation is going to sit alongside what every person has on their PC, on their digital footprint. There's going to be automation helping them along. But exactly what that looks like, I don't really know. I think that's up to all of us to define together and to figure out how to bring automation to the fabric of what everybody does. But it's an exciting journey certainly ahead of us.
Streeter
Exciting stuff. We certainly have a big responsibility. Thank you all for a really fascinating discussion. Really super useful insights on the current revolution in automation and what the future might hold. Thank you so much for your time.
Virdee
Thanks so much.
Smith
Thank you.
Streeter
And a quick note from the legal team. The views of third parties set out in this podcast are not necessarily the views of the global EY organization nor its member firms. Moreover, they should be seen in the context of the time in which they were made.
I'm Susannah Streeter. I hope you'll join me again for the next edition of the EY and Microsoft Tech Directions podcast. Together, EY and Microsoft empower organizations to create exceptional experiences that help the world work better and achieve more.
