Welcome to today's webcast. Before we begin, let's quickly go over how to get the most out of this session.
Your main player area consists of three windows, video slides, and a multifunction window underneath. Each of these windows can be moved, resized, maximized, or even closed if you like. To return to the default layout, simply click the Restore button on the bottom of your screen. The multifunction window is where you'll find some helpful tabs like questions and answers.
An area to download any available materials and speaker BIOS. You can also turn on closed captioning by clicking the CC button in the lower right-hand corner of the video player.
KPMG has been approved by NASBA to deliver CPE worthy training. Today, CPE will be tracking your participation through polling questions, which will look like this. In addition to attending the entire session, you must answer these polling questions throughout the webcast. Your presenter will audibly address the polling question each time one of them appears. Your participation will be tracked and failure to actively participate will result in denial of CPE credits.
If you have any technical issues during the webcast, please submit a question through the question and answer panel and our producers will respond to you directly. Please submit all content questions through the questions and answer panel as well. Thank you and enjoy the webcast.
Hi, good afternoon everyone and welcome to today's webcast entitled GenAI enabled Workforce with Microsoft Copilot for Microsoft 365 sponsored by our KPMG Human Capital Advisory and Microsoft. My name is Craig Hayes and I'll be your host and moderator for today. We do need to cover off to review some administrative items and we have a video on polling.
OK. We'll just move on the video we don't have, but for CPE credit, this session is going to run for 60 minutes and for full participation is going to qualify for one CPE credit. In order to receive your CP credit, you need to attend and remain logged in for the entire session and respond to the required numbers of interactions you're going to see through polling.
CPE interactions may appear either on the top.
Of your screen or within the slide portion of your browser, and it may not be verbally addressed by presenters. Please refer to the Instructions TAB in your media player for any details about receiving CPE credit for this session. If you do need to submit a question, you're you'll be able to submit overusing the Q&A panel on your screen.
Once again, if you have any technical questions throughout the webcast, please use the Q&A panel on your screen.
When the webcast is over, please do not close out your browser completely. The player will automatically refresh to display an exit survey. We definitely would like you to complete that exit survey and click submit at the end of the at the end of the session and then a notification indicating whether or not you've met the CPE requirements will be sent to the e-mail address that you entered when you logged in if you meet both the duration and the polling.
You will receive a CPE certificate that you'll get that in e-mail approximately 2 weeks from today. At this time, I'm going to hand it over to Carla and we're going to get going into our first CPE question.
Perfect. Thanks, Craig. Hi there. Good afternoon, everybody. My name is Carla Baumler. I'm a director based out of KPMG Chicago office in our Human Capital practice. And our first question that we would love to get your guys’ feedback as we're starting the conversation today is how ready is your organization to embrace GenAI? Curious, is it something you're starting to think about or haven't thought about yet; if you're developing a strategy, etcetera because this will help us kind of ground ourselves in our conversation today and how we kind of talk through this.
Looks like we got kind of about what we would be anticipating from responses. It's kind of something top of mind that we're all starting to think about and starting to discuss the kind of where things start to change as you know, what comes next. Where do we get started? Where do we get make those no regret bets or kind of really kind of kick off those conversations, and thrilled for those of you who are lucky enough to say GenAI is in use today and you're off to the races already. So very exciting for you.
We're really kind of going to talk a lot about kind of what you should be thinking about, where you should get started, and what this looks like.
Very exciting, but kind of where we're going to jump into is workforce. So workforce is how work is done in an organization. It's the people. Putting humans at the center of organizations in an AI transformation means emphasizing the importance of culture, collaboration, and personal growth for our workforces. Recognizing AI as a catalyst for your organization's transformation requires an appreciation for the behavior and mind shift journey that you're asking your workforce to.
In the past, technology transformations like SaaS solutions transformed entire end to end processes. However, /GenAI is fundamentally different in that we're actually going to transform hundreds and thousands of different tasks. And this is key, and it's a key difference in how we have to approach GenAI transformations differently because we're actually asking people to change.
How they adopt new ways of working on a task level and by grounding ourselves in this central theme, an objective of enabling our entire workforce, we actually look at transformations differently from an Effective human and AI collaboration to spark this innovation and drive the development we need through this new GenAI technology.
On the next slide, we actually highlight kind of KPMGs workforce dot AI approach. It's KPMGs human centered way of how we are actually realizing the value from our GenAI investments. So this can be used across entire organizations to understand how similar job families and different functions can be augmented or within a specific function.
Next slide.
And so as we actually look at this, it's actually something that we can look at kind of holistically across. But where I want to spend most of my time with you today is actually on the next slide, which is what our workforce dot AI methodology looks like and how we're approaching this at KPMG in this really unique transformation style. So we first start by identifying a workforce opportunity. So what happens there is that's where an organization starts with making their like.
Where all of you are starting now? We want to do something with GenAI. We're not necessarily sure what we want to do but we know we need to do something to stay competitive in the marketplace. And that's where we actually look across the workplace where the most highly augmentable jobs, where there can be the best kind of augmentation of those rules. As we move into kind of the next phase, this is actually where we are implementing those GenAI technologies and actually formally augmenting the workforce. This is where there's a focus on the skills, the capabilities, the adoption styles that are needed for those individual tasks.
As I just mentioned on the 1st slide, one of the key areas that is different in this type of transformation is when we're transforming on a GenAI basis, it's the tasks that actually change. And so therefore, a finance manager today is doing something different than what that finance manager that is also in the role in a different part of the business or something else. And so we look at the individual roles and what they are doing and how they're actually working to understand.
What are the nudges, the activation plans that they need through using a surround sound method and understanding when we have obtained that Capacity savings and actually receive that augmentation potential? That is when we transition into the third phase, which is when we reshape the workforce. In the third phase, it's critical to look at the reshaping of the workforce because this is actually where we are actualizing the value. So this is where we start to pay down the tech debt that we encountered in the first.
Phases as the first two phases were investments in the organization and looking at those technologies where to implementing, ensuring adoption activation, etcetera. And the third phase of reshaping the workforce, that's where we're really looking at the talent strategy. We're looking at the org design, we're looking at how we're further enabling the business strategy that we set out in the business case that started this journey.
Typically, clients I'm seeing are activating new business strategies or enhanced employee experiences. Activating new kind of product areas, et cetera, or others are looking at cost takeout options, but all of that is coming to actualization and really kind of coming down to recouping the value of those investments once we hit that third phase of really reshaping the workforce and that's where we're able to capture the value. The other exciting thing about our approach here at KPMG and how we do this, is while those are kind of three phases that I very quickly kind of walked us through, they can all be kind of done wherever an organization needs to start.
Some organizations have implemented GenAI technologies already and what they need to focus on is the adoption or the learning aspect. Others are just getting started in their journey where we really do need to start out with that workforce dot AI opportunity. All of these capabilities can be grouped together to really kind of make sure we're highlighting and emphasizing the different areas of the organizations that are going to receive the most value through these offerings to make sure that we're kind of partnering with you hand in hand as you're starting down this journey yourself and through the implementation of all of these areas.
I think while I could talk about this and KPMG's workforce dot AI for quite some time, this is where I'd love to learn more a little bit about your organizations again. So we have another CPE. And in this question, I'm curious like what are the kind of key functions that you're anticipating your largest investment? So as you were thinking about this for your own organization, I know a lot of organizations were just getting started based on the last polling question, but I'm curious where you're seeing kind of the most potential within some of your own areas.
Yeah, as these numbers are coming through, it looks like finance and kind of customer IT are really kind of seeing a lot of the focus points and that's really kind of what we're seeing with our own clients. So in a little while, we'll walk through some of our own experiences as we've led clients through their own GenAI transformation journeys. And you'll see that definitely kind of aligns with how we're seeing kind of some of the key areas really playing out through multiple organizations and their own investments.
Rob, did you want to take us through some Microsoft business applications and start walking us through what that could look like for organizations from those business applications? Thanks, Carla. That was a great overview of Work AI and it's interesting to see where people are looking to start their copilot journey. I'm going to spend a little bit of time talking about Copilot and then more specifically trying to address and provide some information on some of the functions specific copilots.
I've been helping clients for over 25 years with front, mid and back office with an emphasis now on unlocking the value of GenAI. Microsoft introduced copilot in March 2023 to provide AI capabilities across familiar Office products. In short, it's an AI assistant that enables natural language chat interface to let users search for specific information, generate text such as emails, summaries, and Word, Excel, PowerPoint, and the rest of the Microsoft Office suite. This enables users to benefit from their capabilities to drive innovation in rich customer and employee experiences, boost productivity, reduce costs, and enable growth for organizations.
As I mentioned, it's an assistant. It's not a replacement for people and it brings the future of productivity so that employees are able to get new tools and skills that haven't previously existed. Technologies continue to change rather fast. Their monthly updates and other features that I'll start to talk through. At KPMG, we've implemented M365 Copilot, and every day I find new ways to leverage the tools to help me with their work. From planning my day, to summarizing emails, to finding relevant answers and documents, or even new fun ways to get a different perspective on projects. And while M365 is incredibly valuable, there are also role specific Copilots that help organizations even more specifically, it was mentioned around finance and sales coming out with areas where Copilot.
These are areas where Microsoft has built role specific copilots. Sales came out in I want to believe February of this year and they continue to make updates to that monthly and then release additional copilots which I'll provide an overview on. These help to remove specific function specific administration tasks. For example, customer service people are able to quickly get knowledge-based answers as opposed to having to pour through the data and files define product information. Finance resources can do various analysis within the copilot for Excel. So they're bringing specific functions and they're done through either icons within the Application or within the tool itself. They can converse with the specific copilots from within CRM, ERP to ask relevant questions and receive answers to replace the previously manual entries.
So these are continuing to come out. I'm going to go ahead and skip to the next slide. We’re going to just talk through several of them. OK, we’re going to start with Copilot for sales. This is a tool that this is a copilot that sits on top of your CRM, whether it's Salesforce or Microsoft Dynamics. It provides sales insights in the flow of work for sellers. For those who have been looking at it in the past, it was used to be called Viva Sales, then it was called Microsoft Sales Copilot, now it's settled on Copilot for Sales. So what this can do is it looks through your CRM Data in conjunction with M365.
To help with meeting preparations, responding to RFP's, summarizing customer histories, and doing company research, the tool lets you update inline within your CRM or within Outlook, so if an opportunity value changes, you're able to get input on how that would change and update it automatically, get contact or content recommendations, and QA for relevant product information and other account communications.
One new interesting feature that just came out in June is surfacing recent communications. So from the standpoint of when you're talking to a customer, it will pop up the recent meetings, the communications, and it can even summarize that information to help people quickly be able to get what they need and spend more time with their clients.
Another copilot is Copilot for service. This is where it sits on top and extends your contact center. It helps users.
Find answers more quickly on a single pane of glass rather than have to pour through warranties, product descriptions, and other help Data. They can actually have it search and summarize to create a response to help them resolve. Now one of the interesting things that I like about recent pieces are when you have a case and it's a new case and there isn't necessarily knowledge based information, you can actually use the case and the interactions with the customer to create a new draft.
Knowledge base for use for other agents. You can also use skill based searching and what this means is I may not be able to solve the answer but I can quickly put in the skills that are needed and search and a resource like Craig may pop up and be able to join and help me resolve the issues.
Now the next two I'm going to lump together customer insights and marketing Copilot exists in both of these and this is really where in the past for doing customer segmentation.
And campaigns marketers and Data analysts would need to write queries to be able to group people, be able to communicate and Target them. Now you can use natural language instead of writing queries to find the answers. This is using a combination of M365 Copilot Customer Insights, Copilot for Studios, and potentially Copilot for sales. Marketers are able to now create content in minutes rather than in hours; Be able to conduct more research and then truly unify the transactional, demographic, and behavioral data. There’s a lot more features coming from that.
The other one that isn't on here that there were a lot of people that mentioned around finance. It is still in preview but Copilot for Finance is a way to access and connect to all your financial systems from within Microsoft AI, Microsoft Excel to do workflow automation, get recommendations and really start to do Data reconciliation. It will suggest reconciliation factors.
It's going to be able to compare Data structures, classify transactions, and soon it will be able to do variance analysis. There are a few others. There are ones within Business Central, which does it helps small and medium businesses be able to market, bring to market faster by helping with product descriptions, publishing information. And there's also a copilot for supply chain center and supply chain management.
That helps to proactively use data and news to prevent disruptions from the supply chain. If we can go to the next slide. We skipped over.
We will ask the question, there is going to be, there is one other question that was going to talk about the Data that came out from a survey, but it looks like that slide has been removed. So we'll go to the CPA question: When will your organization implement its first GenAI technology?
It's interesting to see that about a large percentage or I've implemented to some extent a GenAI offering. But for the most part it's pretty balanced.
Now, I've touched on a lot of the functional copilots of Microsoft's released and there are cases where that may not be enough. And there are also additional tools that I'm going to turn it over to Craig to talk about.
Great. Thanks Rob. And once again my name is Craig Hayes. I'm a partner in our Lighthouse practice and I've, I focus on technology implementation really across the Microsoft ecosystem for the last 25 plus years. Now we have clients that define use cases that require functionality that extends beyond the core features of the Copilots that Rob was talking about.
In the case where you need to extend and create an experience that's more customized for a specific use case there is a recommended process to get you there and that's what's highlighted in this slide. In order to start that journey, really, we highly recommend that you get alignment across cyber, risk, and compliance teams within your organization so that there's an understanding of the boundaries that you're going to operate within to generate and deploy a GenAI implementation.
Now if you're going to on the technology side to start, we would recommend in that case where you need something highly customized that should begin with a privately hosted LLM, a large language model. Azure Open AI is a perfect example. And then while you're doing that, you need to continue to drive to detail the use cases that are going to require maybe prompt engineering cognitive search, something that is outside of the box from the other Co pilots. Once you've established your LLM you need to then connect that LLM with your enterprise Data sources. And it's the Data sources that you want to be able to be ingested and that sets up your foundational plumbing. If you want to think about it that way in your technology. In order to deliver a GenAI experience to users, you do need a development ecosystem where you can bring AI enabled apps to your user commuter and provide a User Interface.
How is your community going to interact with your GenAI Application? And then you can extend your GenAI solution into an ecosystem of copilots that then will bring an interface and interaction prompts that your user community can take advantage of to gain the traction and the adoption of that is going to be. It's critical that you establish an effective like an awareness training, an awareness program and a campaign so that your user community knows how they should use your GenAI Application so they can get the most out of it.
KPMG and Microsoft have been delivering these solutions across our clients. We are able to give you a little bit of insight into our journey and the results that we've seen over, you know, the Data that we have over to the right. We were able to deploy 5 use cases within a three month period and deploy that to our employees very efficiently and quickly compared, when you think about an organization the size of KPMG, that can be daunting. We were able to do that by taking time early in the process to align the firm's cyber risk and compliance teams so they understood the implementation, the benefits, the Data gravity and security that needed to be put in place so you could accelerate a Deployment and then get quick, quick adoption.
As a result, over the last year, we've generated more than 3.5 million prompts across almost 15, 000 unique users. And that's for productivity, apps, content creation and development, use cases. And you can move quickly and you can move at speed if you know where to apply the attention you from an organizational perspective, legal, cyber, risk and compliance, as well as technology. We've seen tremendous adoption in a short amount of time.
Let's go on to the next slide and we're going to talk about another use case. And that is software development. And this is a powerful use case around GenAI and Copilot is gaining traction and developer productivity. So GitHub Copilot is supported across On Prem environments and all Cloud platforms and it's aimed at dramatically reducing your development and testing cycles within your engineering team.
There are four key aspects where we're seeing software development see significant gains and that's analysis, coding, testing and documentation.
So on the analysis side, this is really onboarding teams, preparing, doing the detailed analysis and then a code translation with any legacy code. And when you when you look at the productivity gains, you know that's in excess of 20 to 30% just on the prep activities in the coding side, this is where it's hands on keyboards.
You're generating code. You're going through any optimization.
And this is where you see the majority of your software engineers doing their work, and that's driving 20 to 25% efficiencies. And on the testing side, you know, there's testing automation where testing is being conducted. That's not where a GitHub copilot drives, You know your efficiencies because you'll have different software packages doing that. For testing, this is test code, test case generation and then also brake fix any fixing of coding and this is where you're seeing significant uplift in the development cycles and really the testing cycles in support of testing automation. And then finally on the documentation side, anybody that has done coding knows the value of having context, having documentation and insight into the code that's being developed. And this is where know you have your commentary, any documentation that supports your features and functions that are being deployed. And you're saying a significant improvement 30 to 40%. of improvement in the testing cycle.
So and these numbers that you see here, these are actuals based on recent engagements we've deployed at a manufacturing client. So these are real life examples of the benefits that a GitHub copilot is driving out in the market. And when you add up all these incremental gains for new features and products and applications, it's significantly accelerates the time to value for your users internally and then also your customers if you're deploying new products and solutions.
It's an exciting time for the engineering community to see that their work on a daily basis can be much more efficient and then you can drive accelerated value. At this time, we're going to talk about the client experience journey and then we're going to get into our next CPE question. So I'll turn it over to Carla.
The next CPE question that before we kind of get kicked off on a fireside chat to kind of walk through some of your questions and our experiences with clients, we would love to know.
What is your largest organizational challenge as you're thinking about GenAI within your organizations, you know, curious if it's know impact on the workforce, where to get started, ROI adoption. Would love to know kind of what's keeping you up at night as you think about this GenAI and.
Your own organization's adoption and journey around this space as we're getting answers in, it looks like we have kind of the really kind of key area really being some concerns around Data security and privacy issues, the ongoing governance and as kind of Rob and Craig talked about, that's definitely something that.
Is kind of something they both focus on in both of their areas and really working with clients on because that Data security and privacy issues as you enter into that gene AI space and making sure it's trusted, like there's a whole aspect around this.
Of what that looks like. So definitely that looks to see kind of be the dominant area that's keeping you up at night and with kind of the other areas being kind of relatively equal level of concerns.
I'll kind of jump right in here. So one of the things that when I think about how I work with organizations in terms of GenAI is as I mentioned, we start with a three phased approach. And our kind of first phase is really when we look at identifying the opportunities in an organization. So what we do in that kind of first phase of work is what we're looking for is loosely kind of what I refer to as no regret bets. So where should an organization start to place their dollars an investment on to actually?
Cover where they can get the most kind of bang for their buck as they start to invest in GenAI. So what we do in this phase of work is we take the client or organizational job Data, job profiles, job descriptions, and we deconstruct them down to a task based level. Because as I mentioned, when we think of GenAI, it's really augmenting individual tasks. So what we do is we take those job descriptions, deconstruct them down to the hundreds of different tasks that happen and then really look at them.
In terms of the categories and kind of the weighted value of those activities to understand where there's the most augmentation potential. So as you look at each one of these, you see these kind of key categories have both a Genii opportunity in terms of where GenAI technologies can augment those tasks and enable that workforce to kind of be productive in new ways and where that human work is required. So based on that, what we can understand is the overall Capacity gain in this case for a sales manager for a sample organization. And so it looks like in this organization there was a 36%.
Capacity gain and while that is very substantial, where it becomes really exciting as when you start thinking about this across the monthly opportunity hours, across the salary opportunities and when you take that into scale. So this is really how you can understand the return on investments and how this looks for your organization to actually start making that business case, whether or not the business case is dollar savings or opportunities of Capacity to focus on different kind of new strategic areas.
And really understanding how that works because what this approach using this deconstructed job profile does, it allows us to understand where to get started. It's just a directional starting point that allows us to use Data to actually enable where to start kind of investing in those copilot licenses start to, you know, transforming individual functions. It also allows us to understand where the across the organization there's similar like tasks and activities to actually align what those are and how those can it be enabled by those copilot and other technologies?
And so this is kind of a really exciting place to get started before we start implementing those other areas. I'm curious, Rob, like in your experience with clients, what does this look like in terms of some of the other things that come top of mind? Or maybe there's some key questions from the audience that we want to ask now?
Thanks, Carla.
There are a lot of questions that have come into chat. We're trying to keep up with them and we're going to take a few minutes and address them now. But as Carla talked about starting with the business value assessment, understanding what where you're going to get the benefit is a key piece to start. We also learned in the polls that organizations right now are not ready. They're thinking about it or they're starting to develop their strategy. So we also have Jay on with us and I'm going to.
Read out some questions and we're going to we're having a discussion to spend the next 10 minutes running through that. Based on your client experience, which roller function is usually successful in leading Gen. I implementation in corporate environments? Is it a transformation office, PMO, functional SMEs or other? Craig, would you like to talk about that as far as the governance aspect and then we'll get you your thoughts. Yeah, absolutely that. And the way that you can make this successful is that it's, this should not be a purely technical technology.
Driven exercise, you have to have a governance driven by the business and IT because ultimately your users are the ones that are going to benefit from it. So there needs to be a voice in the room to understand how are you going to implement, how are you going to leverage these technologies for specific use case like finance to what Rob was talking about.
But the IT supports the business outcomes. So governance has to include business IT as well as cyber and risk and compliance, because one of the questions that was in there is how do you protect Data visibility whenever you implement these LLMS, that's one of the things you have to take accounting for is what Data do you want accessible? What Data do you need protected for compliance purposes and security purposes and all that needs to come together for successful Deployment within your environment.
Jay, anything to add?
It's such a great question and it's one that I get a lot right now. Maybe just building on the comments, I think about sort of the foundational capabilities around governance, responsible AI practices that I would encourage every organization to think about at the outset, which is we need to be intentional on where do we think the use cases are for Artificial-Intelligence, that I'm using AI as a broad frame of reference, generative AI as a subset of that.
From there, I think about it in two, sort of divisions, I'll call it for lack of a better statement. First is horizontal work. What are those universal tasks that frankly can be automated and or augmented with generative AI tools. So think about that as communication, think about that as content creation. Those are things that are universal to all of us. I'll say that our knowledge workers, I'll say more traditionally, I'll say bound to an office environment. And then what you've obviously heard is the functional pieces, which is where do we think we've got ripe use cases that are functionally aligned So that could be, that could be customer service.
Generally we see sales and customer service as being the two areas that move the fastest. And 1st for all of us, it probably is, goes without saying, most of us now as we interact with our telecommunications providers, our electricity providers, the first thing we encounter is a bot, which is frankly an implementation of generative AI. And so it's those really being able to identify the use cases that are ripe for automation and augmentation that should drive the decision making.
The last piece of ladder then we can move on, which is look for those things that are, I'll call it labor intensive document management, knowledge management are great examples of this. We're seeing incredible use and interest in adoption from legal both as an industry but also as a function within organizations because they tend to be very document heavy. And we can use generative AI to help on things like summarization as an example and content creation looking for anomaly, so.
There is no one size that fits all, but I think better those two blocks, which is look for the universal opportunities and look for the functional.
Thank you, Jay.
I'm going to try to bundle a couple of these other questions together quickly. And there was a question around a prompt library being used to help staff. So Jay, I know there's some scenario libraries that work for specific things that Microsoft has published. The other thing I'll also say about it is we've been working with clients to gamify and create collaborative environments to share prompts. We do it at KPMG. We have, we use Viva Engage and.
Share props that way to be able to further the use. There's not one-size-fits-all and you're never going to come up with all the prompts and then another one measuring copilot.
If you want to talk a minute or two about Copilot Dashboard and how that enables some of the assessment and usage for sure. So as part of I'll say our Viva product portfolio is a capability called Viva Insights that most of you hopefully will have some exposure to. It's available inside of the team's UI. It provides a core set of capabilities for measuring the change of work and what are the patterns that are evolving as you introduce.
Intelligence inside the organization through Copilot and so organizations that have purchased Copilot for Microsoft 365 actually get the Copilot dashboard as part of that purchase. You have the access to that capability that you can enable. And so that gives you a lot of what I'll call the signal based measurement on how is work changing, how often are people using it, What are the things they're using it most often for, What apps are being activated more often.
But I would also augment that says there's also still space for what I'll call the qualitative a little bit to the comments is what are the sort of the feedback, right? The active engagement. I think the adoption and change management pieces are so important to this because it is truly a rewiring of how work gets done, using a very broad definition of what work means.
Thanks, Jay. Carl, I know this May, this was answered in the chat back, but as far as how to utilize copilot per unit or per field per group, what were the ways that you would start or take the approach with the business value assessment workforce AI to identify those? Yeah, how we kind of started, we really let the Data kind of talk for itself. So once we do this kind of role profile, what we do is as we leave this phase one is we look across.
Organization and look at what are those kind of top roles that have the most potential for the kind of the same areas that Jay highlighted, whether or not they're like the certain knowledge workers or certain functions. Like really looking at the Data itself and letting that kind of really drive where an organization should start. Sometimes an organization has different kind of strategic needs that they're starting at finance or IT for very specific reasons, but typically we really just kind of have this done across the organization and really let the Data be the driver.
And we typically find across an organization kind of those same areas that we just talked about, kind of the key areas where have the most potential are typically across and aligned with the finance, marketing, kind of IT, customer sales kind of organizations and then throughout the rest of the enterprise as well just as aligned to the business strategy. Thank you. And would you use something similar to convince a board or an executive team to invest in Copilot and GenAI.
So when we do something like this, like when we prioritize these roles, what we do is we have all the Data that stands behind it as well. So in terms of what it looks like. So we regularly present kind of a version of this deconstructed role profile in an executive summary format to the CFOCIO because those are typically the people who are the final decision makers of how many Copilot licenses it, copilot for M365 and also Copilot for sales or whatever that makeup of kind of investment looks like and whatever that implementation plan is.
So we definitely kind of use a version of this at the highlighted at the appropriate levels for executives all the time. Thank you. And then one last question, there were several questions around data security and data privacy. Craig, are you able to I mean that that's always the hot button and that's where there are a lot of questions in the.
In the poll, Craig, do you want to hear any comments about that?
For sure and then I'm sure Jay can provide his insights too because it is the most common question concerns about data privacy because whenever you turn these solutions on it will they will ingest the data that it is provided. So if you have a lack of security and some of your Core Data products if you are not thoughtful and plan out how these are going to be implemented, you can discover data that you didn't necessarily want to be discovered and that's where a thorough planning early on in the process can help you define what do you want to be discovered ,and what do you not. And then you can put controls around those data sources and security so that you know you only make available what you want.
And there's a lot of technology that can go into that to analyze, scan and then lock down data that you need to. Jay, did you want to, you look very enthusiastic to provide a little more detail on that. No, you're spot on. And maybe one comment or a couple comments that I'll share. So one which is the LLMS are not trained on any customer data first and foremost.
Second, I'm often, I'll say encountering the feedback or the conversations with customers, which is what copilot is going to find all of the data that I don't want it to find? And again, I'm speaking very specifically for Copilot. The reality of it is that data is there, it's just not easily accessible. And so to the comments is this is really a moment in time for most organizations to get data security right. For so long we've allowed it to just run rampant inside organizations. Copilot is a powerful indexing tool. We call it semantic indexing, which is being able to understand relationships between people, contents, the people that they work.
And it's masterful in many ways, I'll say it, creating those connections. But the data, it does not find data that is not already open and accessible. It just makes it in some cases a little bit easier. And so to the comments made, there are great tools out there, some of which obviously we provide as Microsoft SharePoint Advanced Management is one that we post recently announced to really enable customers to move quickly in this environment to be able to classify data.
But obviously data security is something that I would encourage every organization to have a very clear strategy on, and then be able to monitor it very carefully and closely as you move forward because again, data is everywhere today and super valuable for most organizations. And that's really helpful, Jay. And that's consistent with what we're hearing from clients. We have clients who have wanted to start their copilot journey and they have to resolve their technical debt and for role specific copilots, they need to go through the whole security audit and then the same thing we share. So that's good to hear.
I'm going to turn it over to Jay now. We've tried to answer as many questions as we could, but I want to let him spend some time talking about market trends and insights. Thanks, Jay. No problem. So let me jump in. I'll say just on broad framing, which is the why and I will continue to try to answer questions as they come in. They're coming in really quickly. But let me just start here, which is this is really a new way of working for all of us. It's an opportunity.
Think the way that work has been done for many, many years. And if we go to the next slide, why this becomes so important is grounded in the simple idea that I often talk about is productivity, which is how do we collectively want work more effectively and more efficiently as individuals? Because it is about us in many ways that are the creators and the drivers of what work looks like, but it is also an opportunity to enable and support our organizations and the countries in which we work. And what you can see on this slide, while the data is, I'll say a little bit dated.
For those that are joining from Canada and born and raised and based in Canada, this story holds very true. We've actually just completed another round of research that basically says the Canadian productivity has grown less than 1%, frankly for the last eight years, seven years. And so we have this opportunity collectively as North America and certainly as the Americas to say, hey, how do we go get back to a state where we are driving productivity for our countries, for our economies and then all the way through to our organizations and our individual people.
Underpinning this as we go to the next slide is a couple of realities. And for those that are not familiar, and I can share this link out, but if you want to jot it down, AKA dot Ms. slash work trend index, if you're not familiar with it very quickly, we generally do two major research papers a year. It's about 30, 000 respondents in 30 countries around the world really focus on studying how work is changing. And obviously a big piece of that right now is what is the role and how are we seeing AI start to influence that?
What we've seen in the most recent data that you can see here on the screen is at the end of the day, we as humans, which we are so great at so many things, we just don't have enough time to get the things done that we are being asked to do from our organizations, from our employers. We struggle under what I call this digital debt, just the number of data signals from communications to content that are coming at us.
Its implications therefore, our ability to be able to actually do the strategic thinking that drives innovation. And so when you think about the responses from leaders and leaders and leadership.
Is really this lack of innovation and again, it is buoyed down or weighed down by this sort of digital debt. But what's also really interesting behind this is there is what we called a new AI employee alliance that is emerging that says we as individuals, employees and organizations are actually now getting to a point where we're willing to embrace, I'll say, the machines. And I'm using that as a broad frame of reference, but being able to take advantage of AI and automation to be able to offload that work.
What's also really interesting in the next slide that you'll see that emerged is already we're seeing about 75% of knowledge workers are already using some form of generative AI at work.
Now, what's interesting and behind this, there is still a lot of resistance. A lot of organizations are not yet at a point where they're creating an environment that is conducive to and supportive of generative AI. And so we're starting to see the emergence of this idea of the BYOAI, which is bring your own AI. So very much like we saw it bring your own devices many, many years ago, we're now actually starting to see our people bringing these tools into the workplace, whether supported or not, That creates a huge amount of risk for organizations. And so it's a platform.
And time to be very forward thinking and creating policies and processes and governance models that support AI because our people are already doing it. But what's also really interesting is just look at the difference compared to six months ago, 46% of those started less than six months ago. And so you can see this rapid curve of innovation that's being driven by our people, whether or not we as leaders and as organizations are empowering them as we go to the next slide and we start to think about this broadly as the opportunity that we have.
As leaders and inside organizations that while there is some certainly I'll say fear as it relates to job loss due to AI or job disruption, the reality of it is most leaders today actually worry more about our ability to fill the critical roles in our organization. And I'll just use cybersecurity that sits at the very top of this chart as A-frame of reference today, we don't need to go any further than our favorite news app to just see the exponential growth of security risks.
And cyber security attacks from all over the world and different threat actors. As humans, we are amazing at so many things, but those signals that are coming in from that attack vector, we physically cannot process them. Now add to a reality that says most organizations in most countries are struggling to build the cybersecurity skills for tomorrow. We need help. It's an opportunity that is ripe for AI and for automation to be able to help us sort through those signals so that the limited.
Human capacity we can apply to the most acute of those challenges and those risks. And so that's our opportunity to think about it. You can see a variety of different functions as you kind of go across this curve to say, hey, where do I think about it in the context of the roles that I'm trying to go fill. And back to one of the earlier questions of where do we start? This actually can become a great grounding point for you as you go to the next slide. What's also really interesting and again, the importance in organizations to drive this change and lead from a front foot.
If you can get to the next slide is the data that we're starting to see emerge from LinkedIn. And as many of you will likely know, Microsoft owns LinkedIn. And so our ability to connect to that data source becomes really powerful. And we're seeing exponential growth, 142 times increases in the number of professionals adding AI skills to their profile, over 160% growth, I'll say, over the last couple of last six months alone as it relates to individuals.
Advantage of LinkedIn learning to go build their own AI aptitude. And so this becomes this really magical moment, which is we as humans are looking for help. We're seeing us actually go out and procure that learning ourselves. And so it's a great moment for organizations to be very thoughtful in partnering with HR and learning and development to say, how are we enabling our people to skill up or re skill in this new AI transformation period that we're in. But we're also seeing this dependency, I'll say, as it relates to our this orientation.
The importance of AI skills being part of who we are as individuals. In many ways, it's much like what we saw, I'll say, through the pandemic, where we saw a talent requirements and talent profile shifting to attributes around remote work and hybrid work. And it became a decision point as we were trying to recruit talent into our organizations. And so as you can now see on this slide here, which is this new hiring imperative, which is we're actually seeing leaders actually start to make conscious trade-offs.
That says, on one hand, we acknowledge that those that are earlier in career will likely get more responsibility as it relates to AI. In many cases, those rules may also be the most disrupted because as you think about sort of natural, I'll say organization hierarchy, those perhaps earlier in their careers are more content creators. That is work that we can now lean on AI to help augment. Whereas if we get to more senior leaders or senior positions, we tend to be content consumers and again, a great opportunity for AI.
Help us in things like summarization, but what you're also seeing in the bottom sort of chart of 71% and then the big bold also chart in the center, which is this is actually becoming a defining moment for talent and for hiring. And so for all of us as individuals, as we think about our aptitude to participate in the labor market, we want to go build those skills for organizations. You want to be building those skills in your people so that they can obviously help you drive the transformation. But we're actually starting to see conscious trade-offs.
As it relates to hiring somebody with deep expertise in a functional area relative to their AI skills and making the conscious trade off. If we go to the next slide, what becomes also really important in this, And I wanted to make sure that I hit this one and I'll say up front, which is this isn't just about another tool. I talk a lot about this in the context of mindset, culture and then technology is the third where, in a mindset moment, what I've personally learned is now operating with this principle model that says thinking about sort of AI and copilot.
As your newest direct report or is sent to or shared with me yesterday, think about it as your newest intern, right? The ability to now actually have somebody alongside of you helping you to accomplish that work. But the inference in that is that we still play a responsibility to oversee that, right? As we think about interns coming into our organization, we want to give them great moments to go learn and grow, but we also generally will want to maintain a check and balance that says, hey, is the work up to the quality? Does it answer the questions that we may need, etcetera.
Similarly, I'll say on culture is this is about organizations. Far too often I continue to run into customers where it's like we have a zero AI policy, we are trying to lock it down, we are trying to prevent our people. That is what's fueling that BYOAI sort of momentum. But it is also conducive where those that are I'll say non responsive or non participant, non participatory are at risk of being left behind. So it's a great moment for us as leaders and as organizations to create an environment where we change how we think about these.
And the role they play, we create cultures and environments that foster that innovation, that curiosity, that testing. And then obviously as we think about it in the context of the tools and where best do we implement them? On the very last slides, on the next slide, I wanted to close with this thought. I am often asked as I work with customers across Canada, across the US, across Latin America, I get this question all the time, which is, hey Jay, are the machines going to take over our jobs? And my first answer is no, second part is not yet.
3rd, not in its entirety, but what's intrinsically true is somebody that knows how to use it as a tool absolutely will. We deeply believe, as said earlier, this is a moment of human empowerment and augmentation.
It is not about replacement. There will absolutely be some functions and some work that is disrupted, but I think that is a great representation of where we will continue to transform in many ways. If we look back over the last sort of major industrial revolutions, this is a natural cycle. This is the next wave, the next opportunity. But most important, it's a moment for all of us to participate fully. And so with that, I'll hand it back to the team and I will also try to answer some of the other questions that have come into the river as quickly as I can.
Thank you very much, Jay, and really want to thank the group and thank, thank everyone for joining. There were a lot of questions, apologies we weren't able to get to all of them. Hopefully, we were able to answer some of the highest priority questions before we close. We do have a few housekeeping items now. A replay of the webcast will be available shortly at the KPMG human capital advisory site.
Please share this information with your colleagues who weren't who were not able to join today. When the webcast is over, don't close completely out of your browser. The player will refresh to display an exit survey. We absolutely would like you to take the time to complete that survey and click the submit button. Afterwards, If you met the CPE requirements, you will receive a certificate to the e-mail address that you entered whenever you logged in, and that'll take about two weeks. And then we're also going to get this content out to the group.
After this session, on behalf of the KPMG Human Capital Advisory Team and Microsoft would like to thank you for your participation and thank the speakers and everyone that was able to join today. If you need further information, please feel free to contact anyone of us using the contact information on your screen. Thank you very much and have a wonderful day.
