Welcome to another episode of unraveling IT. And on this episode, we'll be sitting down with Brian Harrison, the CEO, of course, because technologies, as well as Nate Troy, our solutions architect here at Corsica, and they'll be discussing chat GPT versus the newest innovation from Microsoft, which is co pilot. And which one we feel like you should be paying attention to, and how it's revolutionizing the way that we communicate. Stay tuned. Alright. So I'm I'm here today with Nate Troyer, who's one of our solutions engineers. Mhmm. And has worked for Corsica for, well, feels like a long time. I feel like I've worked for you for a long time. It it does feel like a long time. Probably feels like a long time to you. Not not to me, though. It does. But I I greatly appreciate HR you're being here today and working to talk about AI. Yes. So, especially, you know, AI and how it relates to work. And I have a feeling you probably use AI more in your job than than I do. I think I was one of the earliest adopters because I am very lazy. And I hate writing, statements of work. I hate trying to sound professional, and it usually comes off as somebody play acting that they're professional. When AI can just sound professional, use words like seamless and a couple of the hot, hot words that I think, a lot of, c suite people would understand immediately that I am I just have not had that much depth in. And so I, you know, I just started asking, at at the time it was Chat GPT, just started asking it to, write a state write an executive summary, based on the narrative that I was gonna give it. So, essentially, you know, I would just give it the narrative, like, you know, company a, wants to have a refresh of refresh of their data center. Right? It's been it's hold. It has security vulnerabilities. And, you know, company B, that's right. We don't put actual information in there. Company B is going to, you know, perform the work and all that stuff. And it would just spit out the best thing I had ever read that I that did not come from my hand, but was, I would say, enhanced with AI. That's that's the way we put it in our team. Like, it's enhanced. Yeah. And I I think that's a great way to look at it. You know, I I even tell my kids AI is is great for expanding what you maybe thought was possible or the way you were thinking about something. And when it comes to producing work product, it can, it can give us something that that taps into knowledge that that we don't yet have or vocabulary that we don't get half. Yeah. Thanks for invocap. So, yeah. It's really useful. And, like, the intangibles of using it, now I start Now I when I start writing something by myself, just freehand, I have learned how to write well with a I'm spitting that information out. So when I'm engaging with, people in the c suite and I'm writing emails and things like that, I I don't really use chat GPT anymore to write it for me, because I kind of already understand. I might have it, you know, you know, punched up a little bit, but that's about it. So I think, like, it can be used as a tool, you know, to to teach people how to do that sort of thing. Yeah. And and when when I think of generative AI like Chad GPT, that way, even, even, you know, for the types of things that I write, it, it's almost like a mentor. Yeah. Caps into all this extra, extra sources of experience in a way that's so much faster than than just learning it on our own. And I I find that to be one of the things most valuable when I need to think of something differently. You know, we all get tunnel vision get locked into the way we think of the world, and it chat chat GPT specifically in in any AI can help break you out of that. Yeah. And I think since since the, since the amount of information that it's consuming is so great, you know, you can get, answers out of, out of, specifically, like, chat, GPT's model for things like, What are the what is what is the difference in thought process between the average American and the average German? Which is that something I actually actually Because if we're gonna be going into a a discussion with somebody who's, you know, from Germany, they're gonna have a different way of processing what we're talking about than I will. You know, they're, you know, more collaborative and we're more individualistic. So, you know, it kinda helps you, like, it kinda guide you of where you need to go in a conversation. It was, like, super it's amazing what you can ask it. You know, I was trying to think of ways and things to ask it that it probably doesn't get asked all the time. I just saw I read an article the other day that said it's better at doing, It's better it's better at doing, like, research for, like, like, scientific research than models specifically created for that industry three because it's it's consuming vast amounts of data. Right. Which is it's nuts. It's absolutely and they didn't know that until they asked it a question about that they're like, wait a minute. You know, like, it's it's it's giving us better information than the models that we created, you know, for a specific for our industry. So it's it's really cool. I mean, AI is super cool. Yeah. And I I think for for those of you maybe listening that, that aren't using AI yet, you will be. I I think experience is is what gets people to to come back and and realize that you know, it's not a replacement for people. No. But it certainly is an enhancer for your own experience. Right. It's like being able to ask expert anything that that you want to you. And and really, Jeff But you have to understand the answer that it's giving. You know? So there's a component. You have to be you have to be a part of the process though, you know. Yeah. So, generative AI is super useful for business. Mhmm. But we definitely see some risks, and and you started us off with kinda want at the beginning where you mentioned an example of where, where we'll use AI to help enhance maybe something that that we're writing or going present to a customer. And specifically with with chat GPT, there's some risks around, you know, where is that data going? Right? And we live in a in a world where now, you know, I I I see people really willing to give up personal information, maybe accidentally. And, I I observe or or talk with people who I feel like are probably doing that in chat GPT now without without thinking about where's my company data or my personal data going? If you type it into chat GPT or any of these generative AIs, it's their data. Right? So, I mean, you kinda have to be careful. They've written there. What's interesting is they've written their terms of service in such a way that's basically, like, all all risk is on you, which I can understand that. So, I mean, you just gotta know what you're dealing with when you when you get involved with that sort of thing. Yeah. And and I think, you know, take take Bard, for example, where you can drop images in and ask it to analyze those, the the risk goes up pretty substantially Mhmm. If if you start trying to use it to do your job. Right. And especially in an industry like ours, where, part of part of what we do is have data about other companies So it's really critical to to know where that goes, which which kinda, which brings me to to another topic which is co pilot. And Microsoft's generative AI designed around the M three sixty five ecosystem. Now you've done more research on that than I I think you got, like, a test flight of that, didn't you? Yeah. Well, I I've seen a lot of the demos used little bits of it. But it's it's not out yet for small business, which is, you know, in Microsoft's world, anybody who's not a giant enterprise. Well, what do you think so how do you think this is gonna change the landscape for people who are already using the Microsoft ecosystem. Like, what if they're if they're you, you know, guys, if somebody like you is sitting at his desk and he gets an email from somebody asking for financials and stuff like that. Do you think all I do is email? Yep. Pretty much. Okay. Some some days it does feel way. Yeah. So I'll be fired at the end of this, but guys. Don't worry about it. I will never be back. The big differences are that, you know, one, I I don't have to put my information into, you know, chat GPT or or BAR for example, Microsoft is is using those large language models on, my data that already exists in m three sixty five. So it's already designed around my ecosystem. And and the way Microsoft is looking at this is is really application based, but then also broader than that. Through some of some of the chat interfaces. So, you know, I think of that example you brought up, we need some financial analysis on some, you know, maybe in some revenue. Co pilot will will look through my email, and I attempt to gather the data that I need in order to respond to that. And it'll it'll even, you know, potentially draft a response me. That's crazy. Which is just like what we would expect ChatGBT to do. Yeah. It's not integrated. Yeah. Information. And what's great about it it's it's protected. It's it's kept within that space. And it it is a force multiplier four businesses in the right way. Right. Right. So how do you so, like, especially with with our company specifically, like, how do you see it augmenting our our abilities here. Yeah. So it's great if it helps me save time. Right. I think that the bigger impacts longer term are around customer service and how do we interact with our customers? Because that ultimately is our business. There's a lot of other things that that we do or talk about, but we are a a client focused business. We wouldn't have a if it wasn't for those clients that work with us, that we provide security and support for. And so chat GPT does it really solve any of that for us? It it might give us, you know, some some lipstick to put around, you know, a doc that we send out. But what copilot has the potential to do for a company like us is to give everyone the content that they need around that vast array of customers in a really timely and meaningful fashion. And so imagine feeding in who that customer is, who they normally work with, what projects we've done recently for them, what service we've done for them, and then being able to get a summary or a response as someone on the front lines who's interacting with that customer and is able to have that full view. Again, all that data is available now. Yeah. Just we as humans can't go out and in fifteen seconds, go find it all, assemble it, and create some It makes sense out of it. Right. And and give a cohesive response to it. So when I think of what co pilot can do for businesses, search can increase our productivity, but it can really drive us towards a better customer experience, which is what most of us are are after because our customers really drive that revenue and the success of the business. So outside of our our our industry, just outside of IT, we're, like, somebody who's in a factoring job? Like, how do you think that that's going to, I mean, affect sales, affect, the quality of their product? Like, what how do you how do you see it implement it. Yeah. So so in let's use manufacturing as an example. It Because we're in Fort Wayne, guys. Yeah. You know, manufacturing is our is our largest vertical. Yeah. It is. It is. So, we work with a lot of manufacturing companies. So so there's a couple things that that need to happen first one is starting to adopt the data frameworks available within M three sixty five, because that's what's going to bring the information that that we can start applying, co pilot to you. And and so that means that building some connectivity between ERP systems and the Microsoft ecosystem. Starting to bring that data integration into, the the Microsoft three sixty five domain, so that we can start interacting through Excel, through Word, with data from these other systems. And you need people to do that. Right? Yeah. Absolutely need people to do that. You need, you, you need that connectivity but but in a in a world of SaaS applications, everybody has an API. Everybody has, you know, the ability to get that data into something like a Power BI. And that's why, using those external tools to start building those dashboards in in viewing your data, becomes really valuable when you have co pilot, because now you can interact with that ERP data, with that scheduling data, with whatever it is out of those other systems, through a safe AI ecosystem that ties into your productivity apps that you're using all the time. So if I if I was a production manager and I wanted to write a report on scheduling and throughput, if I have those dashboards already built in power RBI, then I can have co pilot help me construct those reports and pull that data those various reports. So, like, one of the things I saw was that you can essentially, since, since chat GPT is, you know, open AI. Right? Right. Which is also behind co pilot. Right. Does is Microsoft do you think they're gonna develop something for people who maybe wanna more hybrid and stay on prem and have, you know, access to their, you know, on prem data? I I think at some point, the in the the connectivity is is there now. It's a little cumbersome to to get it where it needs to be because it it'd have to be accessible through through, you know, what used to be power automate. Yep. And I I don't know. Microsoft is pretty fully invested in getting everyone in, Yeah. No one was surprised that they would have won a pay to play model, but two, their own AI. Right. No. And I I don't think anybody was surprised with that. Yeah. But I I like it as as a business leader. I look at it and say, I can have confidence that I can turn my people loose on AI and do so in a way where I don't have to worry that my my company data is gonna and, you know, on a surface. Well, yeah. I mean, we have we have clients now, and prospects asking us about, you know, what type of policy, Have we what type of policies have we seen out there surrounding AI usage? Like security policies? You know, that that was I I had not heard that just even two years ago. No one was asking that question. Before chat GPT, it just wasn't available Well, yeah. In in your browser to to go and input company data out on the web? Yeah. I mean, the just the the quickness with which this has grown is in is incredible. It's not a new thing. I think it was started, like, nineteen fifty seven. Like, like, into AI is starting to figure out how to process AI and all that. But it really took off when, you know, the graphics processing unit started in play, all the all the hardware that's now involved and how fast it is. And just that everything all that data is now out there for it to consume This is not going away anytime soon. In fact, I think we we have to be wary about malicious use of it too. Yeah. I mean, Right now, chat GPT has some it has some guardrails up in in in case if you wanted to do something, Hey, can you write me, write me some code that will, you know, delete everybody's information off of their c drive? It'll say no. But you can get around it by asking it in a certain way. You know? And so, I mean, I can just see people creating tool sets out of it that I mean, at some point in time, I think it was you who said something on to the effect of one day, we'll just have our AI go talk to their AI, but I think the day is also coming in when, you know, we'll just have our AI defend against their AI. Yeah. I I we're we're likely closer to that than I'm sure it's happening. Realized now. And I think as much of a force multiplier's AI can be for us in in a productive business, it can be in a in a harmful or malicious business as well. And we we certainly see that and and I think if if history has told us anything about threat actors, it's that they're willing to invest time and effort that a lot of times, a a regular commercial business is not willing to invest. And that that keeps them, ahead. Of many of the the small to midsize businesses that we work with that that don't have a partner that's helping protect them. So it it's it's gonna keep growing. And in chat, GPT has guardrails, but there's plenty other ways to engage with AI and produce malicious content. If if that's what your desire is. Right. And, man. Yeah. That's kinda it's kinda scary. That part is kinda scary. It is kinda scary. But I think what I think the upside of this is you're seeing a lot of AI now, being adopted by cybersecurity companies, you know, that have specific products, you know, that it essentially in the background consuming all this information going, that doesn't look right. That doesn't look right. That doesn't look right. I'm pointing that out. And, correct me if I'm wrong, but AI is also in involved in Microsoft's cyber security. It is. It is. Mhmm. Yeah. So I I think we're gonna continue to to see a a bit of battle between Good AI and malicious AI for for a long time. Okay. Yeah. I mean, pretty sure we will too. Yeah. For sure. So I think so co pilot is available for most businesses in in roughly three months. It's available in the enterprise right now. And I know I'm really excited to to get it deployed to some of our customers to see how it can can have an impact in a pause way. But in the meantime, you mentioned these policies around AI usage. I I don't feel like policies or are what's gonna be the answer? Because I I think we all recognize, you know, policies about as good as the paper it's written on. Oh, yeah. Without some enforcement mechanisms. What we're not seeing a lot of now is is companies saying, Hey, will you block these domains so that I can hacked my business because I I think deep down, there's this desire to say, well, I know I'm getting better work product. Because some of my team is using AI. So I expect to see a CS shift as co pilot comes available and and there's a subscription model to get generative AI to to be productive in business. I expect that that we're gonna start blocking some of the generative AI tools from from corporate That's interesting. I did not think about that because they're already paying. Most of these people are already paying for m three sixty five business premiums to like that, you know, that's gonna start moving people away from the the quick and dirty eye. We got a chat GPT account. I it can write my email for me. We'll soak and auto pilot. It's right there. Oh, by the way, it's consumed most of your, your emails and your, in your, in your word documents. In a secure way. And it can give you best response because it knows the most about this specific portion of your company or this specific person who's emailing you. Right. That's really cool. That is really cool. Yeah. It's a great product. Yep. Alright. Well, Nate, thanks for joining me today. Thanks for joining me. You're well, Bill. You're well. I'm the one who runs the company based on, on the vest. Yeah. You're right. You came well dressed today, and I, I appreciate that. So, thanks again for Yeah. No problem. I'm sure we'll have other topics we talk about in the future. Okay. Alright. So are you ready to get started? I'm ready to get started. Okay. Well, I'm here. Who do you think runs the company? Let's try it. Okay. Leave it in.
