Expert Insights on Microsoft Copilot in Financial Services

Hear industry experts, Adam Piercy from Hitachi Solutions and Peter Goth from Microsoft, discuss how financial services organizations can harness the power of Microsoft Copilot and Dynamics 365 for transformative business impact. 

global.hitachi-solutions.com

Transcript

Copy Transcript Copied

Hannah Story
Hello everyone and welcome to the exchanges podcast with Hitachi Solutions.
My name is Hannah Story, and I will be your host for today's episode, where we're going to discuss Microsoft Copilot in financial services. Joining me today as our guests are Adam Piercy and Peter Goth. Adam and Peter, please introduce yourselves.

 Adam Piercy   
 Yeah. Hi, everyone. I'm Adam Piercy.  I'm a Director in the CE practice here at Hitachi Solutions and also focus on go to market strategy around Gen. AI.

 Peter Goth
 Peter goth. I'm a Managing Director and part of our business applications corporate team known as the Global Black Belt team at Microsoft. And I focus specifically on business applications in banking.

 Hannah Story
Awesome. Alright, before we get into the conversation today, Adam, I know you've been a guest on the Exchanges podcast before and Peter, we're so happy to have you joining us today as a first-time guest.  So, I thought I'd do a quick just get-to-know-you icebreaker before we get started.  It's going to be 3 rapid fire questions.
Peter, you’re the guest of honor, I'm going to put you on the spot to go first.

Peter Goth
OK, alright.

Hannah Story 
Now this first question, I got the idea for this, I don't know if you saw it, but I recently  saw a clip of an interview with Bill Gates that was talking about how he would send his daughter a text message to tell her that he was sending her an e-mail and I just thought that that was hilarious and just summed up that the generational gap with his daughter perfectly actually kind of made me feel a little old while I was watching it myself. 

First question, talk or text?

Peter Goth   
With family for sure, it's text first.

Hannah Story   
OK, OK. And window seat or aisle seat?

Peter Goth   
Windows always.  I'm sorry. Aisle seat always. I've got to be out on the aisle.

Hannah Story   
You’ve got to be able to get out.

Peter Goth   
Yeah. Yeah, exactly.

Hannah Story 
Make that connection! OK. And then last question, place at the top of your travel wish list.

Peter Goth 
Probably South America is the one the last continents that I haven't explored. So, South America.

Hannah Story   
OK, cool. And then an aisle seat on your way.

Peter Goth   
For sure.

Hannah Story   
OK. Alright Adam, talk or text?

Adam Piercy   
OK, talk. Even to the point where with my Alexa, I do an announcement to talk rather than text my kids in all their bedrooms. So, talk, aisle for sure, and Japan.

Hannah Story   
OK, you both have long flights ahead of you on an aisle.

Peter Goth   
Absolutely, yeah.

Hannah Story   
Alright, those are great. So, let's dive into why we're here today.

I'll pose a high-level question to you both and you guys can decide if you want to flip a coin for whoever wants to answer first, but I'd love to get your perspective on the value of Microsoft Copilot and business applications specifically to financial services organizations.

Peter Goth
This is something that I talk about every single day and it's a great question, and it's also very, very broad question because when we look at Copilot, I mean Microsoft's got a Copilot for everything. Literally. And so, when we look at it from a business applications, I think I'll start at the top of the stack and that's Dynamics.
 
And when we look at Copilot across our Dynamics space, especially in financial services, we look at areas where we can remove administrative burdens and what that means is just those tasks that don't add value to an individual. 

So, the idea that I can track an e-mail to a lead or an opportunity or to a contact just in a couple of clicks - it just natively works.

It natively knows what to track that and I can quickly get summarization when I'm when I'm working in it. So, I don't have to be looking through and tracking things and that's the idea - whether someone's on the sales side or on the service side - help them do their job so they want to use the system rather than they have to use the system.

So, I look at it from that perspective and then we can go down to the Power (Platform), but I'll throw it over to Adam at the Dynamics level and then we'll come back to Power (Platform) on that.

 Adam Piercy   
Yeah. On the Dynamics level, to your point there, Peter, about saving time, it's real tactical things as well. Like I'm a salesperson in financial services, a lot of it is my last appointment. 

What did I talk about in my last activity? I've recorded that in Dynamics, that information is in the system. But how do I get that out?

And right now, it's a lot of going in scrolling through your timeline. And one of the calls got Dynamics probably recognizes this, opening it up, going back through.

Using Copilot within Dynamics and getting a summary of your timeline.

Just right there on the form as you need it when you need it, I think just that one piece there probably saves probably about 8 minutes a day, if not more.

You know, just real tactical.

Peter Goth   
Yeah. I think on the tracking side, I did have a client who told us - they're in a commercial banking side - the business case for, for Copilot was just in that ability to track emails because it improves the fact that it's so easy people do it rather than having to search and do all those things, so just that one thing, improve the data enough and also save the time for the advisor or the relationship manager that was enough for the business case for it. Those small little things can make a huge difference.

Adam Piercy 
That's the key, isn't it? It's that's it's the small, the small pieces of work where it can just save that little bit of time for very little effort really and implementation, you know is huge.

Peter Goth   
Right, exactly. And it's not getting into that [you, know what?] a regulator or a risk and compliance officer would think is as risky because it's just essentially speeding up something the human would do, and they're still in the loop as far as deciding that outcome and where that should be going. So, it has kind of that double benefit.

And then dropping into if we drop down to the Power Platform, it's sort of a different focus.
 
We'll talk about Copilot Studio, I think separately, but you know, when you look at what we're going and what we're doing with, you know, our new plan creator and things like that, it really is around empowering – an especially in financial services.
I'm not a huge fan of “citizen developer,” I like the term “Business IT” - so it's basically businesses creating things that fall into IT and they may not be developers.

In fact, often times it's better that they're not developers, but these Business IT teams that can rapidly get things through, but the business would normally go back to Excel for or, you know, perhaps access or someone that creating end user computing and all sorts of regulatory issues, the idea that they can control what they can manage it, but they can do it at scale and at speed. 

And we're making announcements - you know, we were just at the Power Platform Community Conference last week and amazing announcements that were coming out with that are that are going to be a little bit later on, but still will dramatically impact the way people actually create applications as we go forward with a lot of traceability and governability and all of those other sort of fundamental plumbing being part of the experience, so tremendous value that we're doing there.

 Adam Piercy   
 Yeah, just on that as well.
 
It's important to remember that these, like you said, these business application developers, whatever the word is we want to use for them, is that anything that they do is wrapped around your policies within your organization with the center of excellence. So, you are in full control of the connections that they can use, the data they can access, how they can access it, what they can do with it.
 
And then going the next step further, you know you can use Copilot itself to help build automation by talking about what it is that you want.

I mean, I've used that myself a lot times. Very powerful.

 Peter Goth   
Yeah, and then if we take Copilot Studio, which kind of stands on its own just from the rapid acceleration we've got, you know, one of our customers is releasing to all of their branches and all of their bankers, a Copilot Studio app that encapsulates all of their policies and some of their knowledge bases so that they don't have to, you know, staff and support a couple of call centers that handle 1.2 to 2 million calls a year from internal employees. So, bankers going to have that so that it's a huge force multiplier for them for one, getting the answers quickly. And two not having to go through that third party to interpret the conversation going back and forth, dramatically reducing costs and creating much more effective experience for the bank.

That's just one piece of it. The other one that we're talking a lot about is how can we improve?
 
You know you've got 10 - 20 different major applications that an individual servicing a client at a bank would have to access in order to do that, whether in the contact center, potentially in the branch. How can we bring that together and instead of like creating a net new application that wraps all those things, why don't we wrap it in a conversation, especially ones that aren’t used very frequently.
 
So instead of actually having to find OK, it's in that program, it's in this screen I have to fill this out and I have to get this done, it’s I'm going to type in what I want to do, and then Copilot will then walk them through in the wizard, like format of being able to do that service so I can keep it exactly the same.
 
 I don't have to be doing training.
 
 It's an amazing new capability that we can then serve, have them service customers again independent of where that customer actually visits them.
 
 So, I just see massive force multiplication on Copilot Studio.

 Adam Piercy   
And I saw this week, Peter the new functionality with the copy and paste into model driven apps. You can take an e-mail that gets sent to from a customer copy and put it into your clipboard and then open Dynamics and paste it in and it will intelligently know where to put it in on the phone on.

Peter Goth   
And it it's well that and that was the same where they actually announced plan designer too where you can literally just describe an application and again this is early, early preview, OK, this is not anywhere close to it's been released, but they demoed it live.

 In fact, they coached everyone in the in 6000, people in the room saying this may not work, but that ability to describe an application and then it would go through and then basically fill out that description to validate - is this really what you're trying to do? And then look for agents.
 
Essentially intelligent workflow that's already done in the background. If it's not there, perhaps create it for you and then do these flows and the flows may not be, you know if they're nondeterministic, they'll jump to whatever makes sense as that next intelligently going through the flow.  But, tracking it end to end so that you know anyone who comes in afterwards can see exactly what happened and why that decision was made.
 
That ability to start with a description and then get all of that coming out of it, it's going to transform the way applications are built.
 We everyone in the in the in the room that gives dead silence 6000 people and the jaws are all on the floor. When they demoed that.

 Adam Piercy   
I wished I was there. All I saw my LinkedIn feed was just everyone just raving about it.

 So yeah, there's one thing that I wanted to just touch on and it goes into everything that we're talking about here is we know that in, in this sector, in financial services, whether it be insurance, whether it be credit unions, banks, investment firms, there is a lot of data that these that sellers - I'm going to put my sellers hat on - that they have to deal with and consume on a regular basis.
 
The being able to summarize documents and compare documents using Copilot and now the new functionality coming out, I mean just in terms of time saving for sellers in a day, I just want to talk a little bit about that and get your thoughts on that and then transporting that data that they have and then have consumed and putting it into a form is a big part of their daily jobs and where they focus a lot of their time and humans hate filling in forms, and we know that a large portion of CRM project sales projects fail because of lack of user adoption, talk to me about how you see Copilot filling in some of those issues and solving those pain points ourselves have.

 Peter Goth   
So it's a broad area for that one. But again, it comes down to that one - is that administrative? I was talking about before, but if I take, remember you talked about taking this unstructured data and then creating structure out of it. That is going to be huge. That's transformative around it.

So, you get an RFP, let's take a large example, you get an RFP that comes in and that ability to just ingest that RFP and start filling in.

You know those things that you would match out of your own content around it? You're now doing a review and you're doing a correction within that.

That's something that you would set up teams for and you would have, you know, project managers and you would go through cycling through, and you do over and over again. And so, this gives that ability to take that unstructured data and create some really, really relevant structure out of it. 

From like an ingestion perspective. Others I want to take this, and I want to make sure that I'm now sharing it. The other piece of this is that - where is that information locked up? If it's not in the CRM record? In Exchange or it's going to be in Outlook or it's going to be Someone's desktop? That ability to take that and make that connection between unstructured and structured data again is another transformation that is going to save a tremendous amount of time and that automation that we could then throw into that.

So as that structured data then gets done Is in launching these automations that would go off in the background and work asynchronously, the Copilot agent capability that's coming, all of a sudden I, you know, a lot of this work and then it's just going to engage me when I need to be engaged in it.

But I also always have that ability to review. I always have that ability to take a look at. Often times, what I'll do is, you know, I have my own voice. I understand my voice and so I will leverage something that that's been, you know created from for me and then I will make it my own.

So, I always have that ability that it's not, you know, something that's going to come out that's going to look like somebody else's. I definitely will alter it to make sure that it's using my voice.

 Adam Piercy   
 Right. And I read that this week that it's about treating generative AI, not as a calculator. That's what Jared Spataro came out and said truly not like a not like a calculator, but like a conversation. And you iterate and you go back and forward and collaborate on what you want to get to ultimately. And I think you know, once you start thinking about it that way with your data, I mean, that's what opens up the possibilities. And gives you more time to be more creative, more strategic to do some of the tasks or some of the things that add value rather than just administrative.

Peter Goth   
 Yeah, I and you know this is way before generative AI came out, Satya had a statement and he said, you know, we need to be thinking that the UI of the future is a conversation, right?

And that's kind of where you were going with that. Is that look, it isn't going to be about “I have this field, and I got to answer this thing, and I got to go to the next and I got to answer this thing. So, some manager can get reporting on it.” 

That's not, that's not effectiveness, right?

This really is around, you know that ability to converse, to create, but also to ensure that you got everything that you need around it so that absolutely I think is going to be the future. It's going to be more of a conversation than it is necessarily a, OK.  I've got to go ABC and then I've got to do something over here in DEF to get something done.

 Adam Piercy   
 Yeah. And it's all about getting that value.

There is this concept of a hawk model around kind of driving engagement, and it is that thing of getting something back that variable reward when you get back the responses and you see what you're getting back from, the generative AI is what that's exciting and that is what?

Then forces you to invest back into it to start getting more out of it and get hooked.

 Peter Goth   
 Right. And the more you use it, the better it gets.

Adam Piercy   
Exactly.

Hannah Story   
I want to ask, I know, Adam, you kind of mentioned earlier you were putting your salesperson hat on and I think in any of the financial services verticals we talked about banking, insurance investment, there's everything comes back to your relationships and kind of managing and enhancing the relationships with the data that you have in in your systems but, I feel like there are so many other roles inside of an organization that contribute to that experience, whether it's a call center rep, or some type of services manager or a Client Experience Officer, it could be any number of roles, and I'm just curious to get both of your takes on, what if I'm in one of those roles or I'm responsible for a group of people in those roles that might sit outside of sales - what would be something that you find inherent to be the most valuable inside of Copilot here?

Peter Goth   
So let's talk about you know, if you go to traditional bank, you've got the front office which is sales, then you got middle office which is typically handling you know the risks and then all of the back office which is handling all the stuff post right all the operational pieces around it, and if you look at the amount of work that they have to do to carry something from one position to another, so whether it's, you know like a loan origination and I got to go to underwriting and then I go to approvals and I've got to go to all of these different account opening right fulfillment. 

And typically I'm handing it off and someone else has to kind of bind it and do all the work automating those pieces so they can focus on the things that they want to do rather than the things that they just need to do in order to keep this thing going.
 
 I think it's going to be part of that and then having built in intelligence around if something falls off the rails, so if something doesn't, match or somebody is off, you know, family emergency or something could happen and they were handling a bunch of things is to be able to automatically, you know, have it handled by somebody else around it because that's there's so many disruptions that occur in these processes and the more you move up in the complexity, so you go to commercial and corporate banking and you go to capital markets and that they get more complex but they also get more even more regulated around it, so there's more oversight required and more risk and compliance pieces that come into play.
 
 Just making sure that the guardrails are there, not necessarily for the AI and obviously that has to be there by nature, but for the work that's actually being done is just going to create and it's going to be a multiplier of effectiveness and efficiency that banks are going to be able to take huge advantage of and it's not going to be replacing people necessarily. It's going to be making them do again what they were trained to do rather than, you know, having to do data entry for the sake of data entry to make sure that it gets somewhere else. Adam.

Adam Piercy   
I just wanted to just to pick up just a little bit on that of what you said is you know today we have CRM systems that do this and we have automations and it passes them through and we have approval gates and we can build that process, but I think the secret here is where does Copilot generative AI fit into that process?

To stop rework to speed up to automate and to enhance ultimately in those processes and it is about the carry through of the data. It's about identifying some of the you know, potential flags. It's about serving information at the time people need it and recommending it to them. And it's about not losing it in your inbox if that's where it goes or losing in your Teams. And it's about collaboration and getting answers at the time you need it in order to complete your work on time to move it through the gates.

So, I think it's a big piece around the whole platform all coming together to sing harmoniously, to drive that experience.

Peter Goth   
And actually, just to kind of throw one back at you on that one, is that it also should reduce over engineering because that's the one of the things is the edge cases that kill, right? It is where somebody engineers something for every single case. 

 If it's done for the, you know, let's say the 80/20 rules, done for 80% of the work and the 20 is done with edge cases, that allows a much more streamlined process.
 
And then it allows the intelligence on the edge cases to remind people about things that have to be done for those edge cases, right? Because oftentimes they're not done.

Frequently, they're done every so often, so people they're doing a ton of research to figure out how to handle those things. And so I think, you know, that's one of the big impacts this is going to have because every time we try and implement a fairly sophisticated flow for the back office or middle office in a complex environment, they want to say, well, we've got to handle this. 

I'm saying, OK. For now, let's manually hand it off.

But going forward, that easily could be a Copilot and intelligent assistant that says, hang on, this isn't going to match what you need, or this is the next step you need to do.

That is going to be hugely transformational in these processes again, so that the stuff that's not that 80% just flows the rest of it again, people can focus on, but they've given the assistance to focus on those edge cases.

Adam Piercy   
I love that Peter, I didn't think about that, that's brilliant. That's a great idea.

Hannah Story   
 All right. I'm going to pivot slightly. You've mentioned it a couple of times, I guess in in different ways, but I think you can't have a conversation around Copilot or Gen. AI in in broader terms, when you're talking to any of the regulated segments without talking about just that - regulatory compliance. Ethical considerations. Maybe there's bias here.
 
I think there’s trepidation and there are legitimate challenges that these organizations are facing with wanting to keep pace with innovation and bringing these amazing tools into their organizations, but rightfully so, may have some alarm bells ringing off in different departments.
 
So, I'm curious to get your perspective or your take on that, what you are hearing from your customers or what we're saying in response?

Peter Goth   
I guess the first thing I'll say is that since the regulators are really not stepping into this world anytime soon, right? So, what happens then is when the regulator's not saying anything, risk and compliance will say, then nothing happens like we are not going to be at risk as a bank when we don't have a position from the regulator.
 
We are working with some of the GSI’s to try and see if we can get the least opinions and some of those others in place.
 
But realistically, you know this is this is a one of those transformative moments. And it is, you know, if you're not taking some of the crawl steps at this point, you're going to quickly, quickly be well behind around it.

And it's not necessarily something you want to jump into when you know everyone's at an accelerated pace around it.
 
It's just really, really hard to catch up with. That being said, there is, you know, ways that you can crawl in this space around.
 
You know if you know we've I think we've got one customer globally that's using Copilot in a customer facing scenario, but, you know, the vast majority is keeping it internal. So, let's use it for our people. Let's help our employees be more effective. We have eyes on them, and we can explain to a regulator because the other side of it is it can be all tracked.

So, we can explain to if somebody from risk or somebody from compliance or the regulator comes in and says well, why did you make that decision? You've got that flow around it; you've got that support to be able to say that, but you know by keeping it internally and using it internally. Huge advantage to that. You know the ability to reply to an e-mail that comes in from a client, you track it, then reply to it.

You still need the eyes on. You need to make sure that your people are actually coming in and you know, adjusting as to again voice and things like that so it doesn't look like it's Copilot that's returning it.

So, the other pieces on the ethical side, I mean, we're doing a tremendous amount in business applications.

We actually have pre-prompt and post-prompt processing to you know reduce any inappropriate language that that could be ingested or coming back. Reducing the chances of hallucination and things like that. So, and the fact that it is running against your data rather than running against web data is a huge, huge advantage to making sure that one it's your content and two it is it's got those guard rails that you already have in place.

Adam Piercy   
And there are more guard rails that we can put in place as well, right, Peter?
There are things with the Azure AI Content Safety Studio that you can put in place.
Prompt shield ground-ness detection that further enhance that and give you a bit more control.

I'd also say that where I've seen successful AI adoption within an organization is where they put in place an AI Governance Committee of people who are understanding of the technology, not just afraid of it, but understanding enough to ask the right questions. Define the use cases. Define what the criteria are rounded and what the guidelines are and what your ethical guidelines are, and then put those things in place in order to have a, you know, some rigor around the AI and you know, just kind of opening it up to the world.

 Peter Goth   
Absolutely. I mean that should go along with any of your existing governance controls as well that you're doing, you know, anything within the cloud for example, or even just from a systems and perspective internally that should just be one of those, those groups that provide an input and to be honest with you risk and compliance should probably be sitting on those committees right to be providing feedback and engaged early.

Adam Piercy   
100% Yes.

 Peter Goth   
I did a session at the Power Platform Conference with both bankers and they said, look, we invite them in early to what we're talking about doing, and it makes a huge difference because we're not coming to them with something that's, you know, maybe not fully baked, but close to fully baked and they say, well, you know, we need to understand this better and or we may need changes around it at that stage and it's incredibly expensive to do those. So having those teams engaged early on can make a huge difference. If they're part of those groups then, I think that's just a natural place for it, and you'll find that they go from saying no to saying, let me give you my feedback around it and it makes a it makes a tremendous difference in in the outcomes that you've got out of it.

Adam Piercy   
Yeah. And I would also say as well and nothing just gets not forgotten about, but not talked about enough really is just because it's I mean because it is generative AI, there is no harm in having human oversight. It doesn't mean that the AI is failing. But having human oversight to go in and verify the AI is the same as me going in and verifying a solution consultants work. There's no harm in it.

 Peter Goth   
Right.

Adam Piercy   
It's perfectly fine for us to do that, and you should.
 
Peter Goth   
Yeah, one of the side conversations I had in that session was talking about some of the, the regulatory requirements that are out standards that are that are that are coming around AI and one of the interesting pieces is that AI needs to test AI.  which is an interesting, you know concept and I think that would on its on its face would probably concern a regulator or perhaps risk or compliance.

But the idea is instead it's not the same model, right? So, it's going to have, it potentially could be expecting different outcomes around those pieces, but you can do it at scale so you're not depending on someone, saying OK I've got to assimilate all of this data to try and see if the outcome that it that it gave me is correct or makes sense.

It allows for that scale capability to occur when the AI differentiates doing, you know whether you want to call it unit testing or just overall AI testing and I thought that was kind of a fascinating discussion.

 Hannah Story   
All right. I have two more questions before we wrap this conversation today. I know nobody has a crystal ball, although maybe that would be nice - or not if you don't want to be able to predict the future - but Peter, from your perspective, where do you see the journey of Copilot heading from either inside the industry or on a broader scale?

Peter Goth   
I think I'll take it within the I guess the framework of the broad of the industry with a broader kind of mind.
 
 I think conversation is going to be the kind of where we're going with this and that is it's not you know you're not going to necessarily have to structure anything because that structure may not be the best way of doing something and the AI will be able to evaluate multiple paths and get something done a little bit more efficiently and effectively and effectively and then AI bringing the human into the conversation when necessary when things don't seem to line up or when it needs some human intervention as we go forward.
 
I just seeing that getting better and better. So that again it's the idea is to create efficiency to create effectiveness, have the human in the loop so that we're, you know they're doing. But all of that stuff that that we all kind of grit our teeth to do -I'm working on some stuff right now and I wish I had a Copilot working with me on it right now, preparing some reports - but having that ability to say get me this data, put it in this format and do this or let's take this, you know loan package and let's get it to the right people so they can review it, but make sure it's in a format that that they'll understand, right?

So, you know, breaking out the financials, grabbing data from here, doing the market analysis, all of that stuff that somebody has to you know, drudgery for, for lack of a better term, that they can then just focus on what they should be focusing on, what they've been trained to do and that's do the evaluation that they that we're supposed to. I just see that getting better. And you know the idea that we can, you know, be more effective allows us to do more.

I don't see us sitting back and chilling right? It's not going to be the Jetsons, but I just see that getting better and the, you know, the intelligence just, you know, taking, you know, steps forward to help us in what we're doing and keeping us on that path of ethical and responsible AI.

 Hannah Story   
I love that, and I have to give you a shout out for the mention of the Jetsons. I appreciate that. 

Adam going to pivot a little bit on the same question - what are you most excited about?

 Adam Piercy   
I am most excited about the need for forms and filling in forms as human beings, we've been filling in forms since the days of the Magna Carta and I think we're finally getting to a point where we don't need to fill in forms as humans anymore.

And I think that for me, that's so exciting as someone who immigrated country, immigrated countries and had to fill in just millions and millions and millions of forms in order to move countries this the idea that an AI bot I could just tell it what to do and it would fill it in me, that's it to me.

Hannah Story   
Note to Self, Adam does not want to fill out forms anymore, got it.
 
Adam Piercy   
That's a personal point.

Peter Goth   
Yeah, so let me add in my - as everyone's kind of talking, I've been kind of in the back of my head going OK, where do you (Adam) see as someone who builds things right? As someone who architects things as someone who configures things and you may write code, I'm not sure, but as someone who does that as someone who like I started programming well over 40 years ago in COBOL on punch cards, where do you see that affecting your work? What you do for Hitachi [Solutions]?

 Adam Piercy   
That's a tough question. I think, as system implementers, you know, I still think that what we do is we build relationships. We build relationships between data and I think that we still need to be able to do that and we still need to know how to explain and rationalize what those are, but I think the days of us going in and pointing clicking a mouse to do that and typing in the names of the fields, I think in a few years we won't be doing that anymore.

Like the actual fingers on keys doing some of that work but the commute they're talking to it and I kind of see there's a manager/subordinate kind of relationship - I think we all have to become managers of our own AI and then we level up through our management.

Peter Goth   
I'll add one thing to that because I think you're right, but I think it's the creativity piece too it is. That is that unique side of being able to envision. And I don't do it in in real time because I'm a I'm a design thinking proponent and I use a lot of design thinking in in what I do, but I'm kind of as I'm going through this discussion with a client around what they're looking to achieve, there is kind of OK if we you know, if we put this in a big picture layer this is how this could, and so what you're what that ends up becoming is that creativity becomes a description, right?
 
That creativity and that I think is where it's going to move towards.

 It's less about again, like you said, hands on keyboard and more about really thinking about what if we could when you need to do this, this is what happens right and getting into that ideation. That's also where I think it It's going to go.

Adam Piercy
Yeah. I think that's great. I do design thinking myself, and in fact that's where I kind of point the term “Invisible CRM”. I was like, hang on, CRMs are inherently hard to use because it's filling in forms of putting data.

And so, I came up with this concept a few years ago - Invisible CRM. It was all around this concept of taking unstructured data, making a structure, putting it in that I think we're moving towards. We're always thinking about there is well, I see the role like. So, I think about myself and some of the roles that I do with Solution Architects, a Solution Architect's job is to say – “So what?”  - it’s to do the big picture creative thinking piece. But then So what? OK, we do that. So what? So what? So what? And how do we go down now with, you know, AI helping their journey, helping customers as well cause, you know, they're not going to have all of that. 

Speaking of customers, my customers who I work with every day who are purchasing Copilot is there is a skill gap that we need to bridge between from what we do today to where we need people, where people will be using AI and how we talk to AI, that skill gap and how we help organizations on that journey as well. I see that as well as, say, Hitachi solutions.

Hannah Story   
Great. Last question - I guess you could take two approaches to this, and it might be the same response, but either a key take-away from your conversation, something we talked about that stands out the most to you, or your biggest piece of advice to someone who is early on, or still evaluating where they want to go with Copilot.

 Peter Goth   
I'm going to start with a piece of advice, and I've said it once before in, but I think it's worth repeating. You know you want to go; you need to step into this world, but you don't have to jump in with both feet. You don't have to hit the deep end. You can, you know, go in lightly, get up to the knees. Make sure the water temperature's right. But really, start experimenting with this. Do it in a lab. Do it in an environment outside of what you're doing.
 
Understand you know that this really should be an internal impact before you ever consider going outside, I had a client came to me and said, oh we liked and I said you're risking compliance teams just never going to allow you to do that without, like, three years of serious discussion. 

But then getting in the water soon, because unless you do like I said, things will start accelerating. We can see that that we're in the beyond the curve of the hockey stick. Now this is the time to be again.
 
You don't have to jump in, but to get your feet wet and start exploring this because the benefits are real. The acceleration is real, and you know there's nothing like experience to be able to inform what you're doing. 

And to be blunt, engage a partner, engage a Hitachi, engage a partner in that discussion to help guide you along the way to get the right Copilot - Power Platform Copilot, Dynamics Copilot, whatever that might be. Engage that so that you've got that advice. You got that ability as you're in that early stage of helping you stand up and holding your hand as you go forward with it. So that would be my advice.

 Adam Piercy   
And that's great advice. I'm not sure I can add much.

What I would say is that what we've seen is that it's very easy for us to define business value and to get the business value for your organization and prove it. We can help you with that. 

I would also say is to check your licensing now, because you might have some quick wins that you could start using AI with. If you’re using Dynamics and you've got Sales, you could probably switch on Copilot today and start seeing tangible benefits immediately just by getting Copilot summary of your timeline instantly without doing very much so that would be my piece of advice as well - start small, go.

Hannah Story   
I love it. Well, Adam, Peter, thank you both so much for a wonderful discussion today.

 I want to thank our listeners for joining us on today's episode and let remind you that you can subscribe to the Exchanges with Hitachi Solutions podcast wherever you get your podcasts or find us on our website at global.hitachi-solutions.com to learn more. Thank you so much.