Exploring the Future of AI and Custom Copilots at Microsoft Ignite, Seattle
Posted on November 17, 2023 by Fusion Connect
Watch & Listen
Tech UNMUTED is on YouTube
Catch up with new episodes or hear from our archive. Explore and subscribe!
Transcript for this Episode:
INTRODUCTION VOICEOVER: This is Tech UNMUTED. The podcast of modern collaboration – where we tell the stories of how collaboration tools enable businesses to be more efficient and connected. With your hosts, George Schoenstein and Santi Cuellar. Welcome to Tech UNMUTED.
GEORGE: So, this is a live episode of Tech UNMUTED from Microsoft Ignite, Seattle.
SANTI: Literally live. We are in Seattle. We are together. By the way, I am freezing.
GEORGE: And I am not.
SANTI: He told me let's go record a podcast. He didn't tell me it was gonna be outside so, but this is great. We're in Seattle, where at Microsoft Ignite, we're alive and the energy here is phenomenal.
SANTI: Amazing energy. So, what did you hear today?
GEORGE: So we heard a lot of stuff. There were a lot of chip announcements, AI enabled stuff, Data Center stuff right? That was pretty cool. The thing that really came through was the different things from the launch of Copilot.
SANTI: Oh yes! Yes, yes. So, one thing that I heard that was a big takeaway for me was the whole ability to create your own Copilot using what they call Copilot Studio. Imagine now having a GUI driven interface, which is a low-code interface (we all know about low-code) to generate and create your own Copilots that are grounded in your specific data for various specific needs. And I saw an example where they actually were taking kind of like a virtual agent flow and the screen and creating a Copilot for that specific virtual agent. Phenomenal stuff. And it is not just Copilot. They are also allowing you to do AI Studio to create your own AI modules.
GEORGE: And the key on this, this is enabling businesses.
SANTI: Oh yeah.
GEORGE: But it is not just the business as a whole. Like IT creates this thing and hands it off (although that is one of the use cases). It allows individual users to go in a really low-code environment to enable their daily work better. One of the key things – because there’s a lot of questions like should I pay for Copilot? The service is $30 per month. There is a study that on average, people are saving 10 hours a month by using Microsoft Copilot.
SANTI: Oh I believe it! I mean, did you see some of the examples they had? How people are plugging in different prompts? I’m like, I want to use that prompt! I want to write some of those down. Especially on Excel. The Excel one is phenomenal. Think about how many times you’re trying to figure out a formula.
GEORGE: Oh yeah. Exactly. Or pivot tables. I can do pivot tables, but it takes me forever to set them up.
SANTI: Did you see the one example they used of an electric car company and how they can build their own Copilots to alert them when a charging station needed maintenance? And all of this process was automated through their own Copilot. That stuff is mind blowing. That tells you, George, that your limit is basically your imagination at this point. You have the studio, you have all of the elements in Azure, what do you want to build? And if you can figure that out, there’s nothing stopping you at this point.
GEORGE: And even beyond just the builds, they did a couple demos on different capabilities of Copilot directly into your daily work. One of them is plugged into a lot of the standard interactions you do. So, my initial impression was, hey, it's sort of like GPT plugged in and it looks at an email and says, hey, why don't you respond to this. But would it actually does, is it looks at everything that you do and how you interact.
SANTI: That's right.
GEORGE: And it makes the email sound like you.
SANTI: That's right.
GEORGE: Which is the most amazing piece.
GEORGE: Which is where the savings really comes in.
GEORGE: We’ve talked about Teams premium before, which gives some automation as an example, in meetings where it's tracking meeting notes and a couple other things, but it's post production.
SANTI: It's creating AI generated notes and creating suggested tasks.
GEORGE: Exactly. Yeah, but now with Copilot running against the meeting, it actually does it in real time. So you can see the meetings being created. You can correct Copilot and say you missed that task or comment that Santi made and build those into the meeting notes again in real time. You can actually go back in after the meeting now with AI and ask it queries on what happened.
SANTI: That's right.
GEORGE: So if you missed it and you have specific questions, you don't have to search through, you just ask the point.
SANTI: So listen, they made very powerful statements today, huh?
SANTI: First of all, did you notice the use of the term Copilots and plural?
SANTI: It's no longer just Copilot. It is Copilots because now because you're able to build your own, they made it plural. We can have Copilots everywhere. They also said something that blew me away. Did you hear the whole the world's computer?
GEORGE: Yes. I put that in a post today actually on LinkedIn.
SANTI: I love it! So basically, they've built the world's computer and so Azure, right, being the platform for that. They're really, listen, we said it from back in May when we were in Vegas. And they announced it and we said that they're throwing everything at AI and everything at Copilot, didn't you hear that today?
GEORGE: And it is Copilot.
SANTI: It is Copilot.
GEORGE: No, it's interesting because I was at Gartner a couple of weeks ago, Gartner Symposium, and it was AI, AI, AI. It is that here, but it's in the form of Copilot.
SANTI: That's correct.
GEORGE: It's your effectively your personal and your business assistant that's really going to enable what you do.
SANTI: 100%. What powers did it behind the scenes? Did you see all of the partners that are involved in building these massive machines? Did you see the compute power?
SANTI: The compute power, so those of you who love computers, your head would have spun today because the chips, new chips, racks, rebuilt racks with amazing cooling systems for AI purpose-built data centers. All this was announced today it was like you couldn't keep up with it. It was amazing.
GEORGE: That's a drive for efficiency, right? The speed increases with the chipsets. Massive. 47% like generation over generation increases. The real focused by Microsoft on environmental responsibility, both in chip production and the power usage of the chips. And then there are actually sources of power. They're effectively trying to be zero carbon by 2025.
SANTI: Which is outrageous when you think about a data center. How do you pull that off? They're doing it. They're on track to do it by I think it was 2025.
SANTI: So anyway, we're just getting started. Like we're only a couple of hours into this and we're going, dude, how do we keep up with this stuff? And this is just like, honestly we just started a couple hours ago, the information that's flowing is very fluid. It is significant. AI is the driving force. There's nothing else being talked about other than AI and Copilot, and how literally it is changing not only how we do work, but it's changing the world. It really is.
GEORGE: So we're gonna close this one out. We'll come back probably with a couple of different things over the next few days. I think maybe we'll sit down again and this afternoon and talk about some of the conversations you've had. What those sound like and we'll see what announcements come out tomorrow. But take a look at this one today and we'll be back again soon.
SANTI: Stay tuned. We're not done. We're just getting started. See you all soon.
GEORGE: Welcome to day two of Microsoft Ignite in Seattle. Another live episode here of Tech Unmuted. We want to talk today about a little bit of what we heard in the the keynote today. Couple things stood out for sure. So, first thing was Copilot Studio. So Santi, I will let you hit on that one a little bit.
SANTI: So Copilot Studio is basically a low code platform where you can develop your own Copilots right? And right away from a marketing standpoint, my light bulb went off. I know yours went off. There was an ability for us to add a Copilot where we can tie it to some different sources. For example, HubSpot or Salesforce. Now we can ask questions of Copilot inside of Teams and guess what it’s doing? It is tying through APIs to that Copilot to draw a generative answer on those sources. Whether it is HubSpot or Sales Loft or Salesforce.com. So man, this is great because think about it, our team can now ask Microsoft Teams Copilot, “Hey, can you tell me when was the last time we contacted this customer?” or “What exactly was the customer asking for in the last meeting?” and it will go to the notes and all these sources and generate a response.
GEORGE: Or “What are all the touchpoints?”
SANTI: Yes, exactly.
GEORGE: Although all these things are integrated, they talk to each other, they pass data back and forth, you have to be able to dig. I mean in Salesforce alone, you can go through multiple tabs on a contact record, within the contact record a bunch of detail about what the proposal was, what conversations were, all those strings can be aggregated with Copilot. And get a complete integrated view with control over the way the data is presented back as well so that you get a consistent response. Often times if you use regular GPT-4 today, you get a different response any time you ask the same question
SANTI: Correct. So, I really like that. I think that is a project we should take on as soon as we can. We could, for our sake, generate our own Copilot, so I am really looking forward to that. Think of Copilot Studio as another Power platform, right? Same concept. I can create an app, I can create custom workflows in Power platform. Ok, great. Same exact idea, but now I can create my own Copilots. And they have API where you can tie into different data streams. One of the things I found really helpful was, in the Microsoft world there is all sorts of different data bases or type of data right? You have things like for example, the data verse, you have graphs, you have lakes, you have all these types of data and they all do something different. They all live in different buckets, different places. Microsoft just released, well they had it in preview, they announced it now it is in GA, Microsoft Fabric. What I did not realize, though, was that not only can I tap into these different types of databases, data queries, within the Microsoft ecosystem, but I can tie into databases outside of the Microsoft ecosystem. Including Amazon for example. So I can grab an S3 you know bucket from Amazon with data and tie it into the fabric and what fabric does, it brings it all together so now I can point Copilot to that and get me generic answers based on all that data.
GEORGE: And more quickly, right?
GEORGE: Quickly with a set of controls around managed security and compliance and other kinds of things from a data standpoint.
SANTI: You can also go to advanced data features in the Azure cloud to further enhance your encryption. But the fact that you can take the Microsoft fabric, which is now available for general consumption, and tie in all these different buckets including external buckets and then tie that to your custom Copilot you just created... I said it yesterday. The limit here is your imagination. Like, right now, and I don’t know about you, but I feel like I am in the driver’s seat.
GEORGE: Oh yeah. So, there’s been a shift. The push on all of this today has been around low-code, a new term I heard today pro-code...
SANTI: You know what that is right? It is traditional programming. Now they just call it pro-code. You have low-code for dudes like you and I. We’re not programmers. But when you need actual programmers step into developers, pro-code. I heard that too. That is a new term now.
GEORGE: And there were a couple more things. So you mentioned pure view. Pure view gives you the ability to set a bunch of controls around data. How it is used, where it is used, it stays persistent around where the data is used. Once you are within the Microsoft environment, you change the way that you interact with data and you reduce the level of integration effort and other things. Another thing they said today was the CIO wants to be the Chief Information Officer. What the CIO does not want to be is the Chief Integration Officer. So you do a platform play and you get out of that requirement to be cross integrating, worrying about the how data is used, worrying about the cost of multiple different platforms integrated together and then not really having the visibility across that data that you would want. You never get to when you do something other than platform.
SANTI: That's correct. I love the the Lego story.
SANTI: Yeah, so Lego, of course, had everything you know, they were in control of everything on Prem. Everything on Prem and they saw that, hey, moving to something like Azure platform. Also, not only can they can they expand and scale, but they can do so much more now because they have tools at their fingertips versus trying to figure on their own. It was counterproductive. We've known this for a while. That an old approach to technology.
GEORGE: It's not efficient. It's not resilient. You can never make it.
SANTI: You're never gonna be up to date on technology. Look at Microsoft. You can’t keep up with them. So if you are already part of the Azure Cloud, you’re benefiting from all their innovation. Why wouldn’t you want to be there? To me that makes perfect sense. There's so much going on. These last two days, it’s been just information packed. Like, my head is still spinning from yesterday. Today was a little different than yesterday though, right? Yesterday was very much about the end user being in power and the end user getting involved in Copilot. Today was more of a developer’s perspective.
GEORGE: And data integration, how do you work across data sets. Well the last thing we had our list, we want to talk about today is the Azure AI content safety which is the ability to put a lot of controls around what is presented. You would potentially put more controls over data that goes externally in your organization, maybe a little less on what's used within your organization, but again, you can control the presentation and get a much more consistent outcome than you get again with something like a GPT for today.
SANTI: Yeah. So it's funny you should mention that because in Teams we now have a few new meeting options.
SANTI: And so one of them is actually the Town Hall, which is replacing the old live event option that was there. So that's live events going away, but there's one if you look carefully, there's one in the drop down now called Control Content. It is a preconfigure meeting where you can have certain safeties and policies around the content about to share including watermarks and all. And all of that's obviously being driven by Azure and so they were smart about it, they basically created a template for you already. So if you're gonna share very sensitive information, you would pick that and schedule a meeting on that. It's really cool.
GEORGE: And it's sensitivity, one of the interesting things as well was it's even about copyrighted material. So it will be able to strip out and identify stuff that's copyrighted and keep that out of presentation that goes back.
SANTI: I find that fascinating.
GEORGE: So hey, with that, we're going to close it out for today. Beautiful day.
SANTI: Yeah, much warmer. I am not cold today. It's nice and cool. The sun is out. The sun's out. In Seattle. It's beautiful. This is amazing. Listen, Seattle, such a beautiful city. I've never been here, first time. I'm enjoying it. I'm looking forward to spending more time here, folks. We are gonna bring this podcast to an end. Not because I have a producer tell me to do so, but because we have so much to do. Like we're just getting started with day two. But listen, this is a good time to remind you to subscribe. Subscribe to Tech Unmuted. Do it right now. Just subscribe on your favorite podcast platform and until next time, remember this. Stay curious, stay connected.
CLOSING VOICEOVER: Visit www.fusionconnect.com/techunmuted for show notes and more episodes. Thanks for listening.
Produced by: Fusion Connect
Tech UNMUTED, the podcast of modern collaboration, where we tell the stories of how collaboration tools enable businesses to be more efficient and connected. Humans have collaborated since the beginning of time – we’re wired to work together to solve complex problems, brainstorm novel solutions and build a connected community. On Tech UNMUTED, we’ll cover the latest industry trends and dive into real-world examples of how technology is inspiring businesses and communities to be more efficient and connected. Tune in to learn how today's table-stakes technologies are fostering a collaborative culture, serving as the anchor for exceptional customer service.
Get show notes, transcripts, and other details at www.fusionconnect.com/techUNMUTED. Tech UNMUTED is a production of Fusion Connect, LLC.