Catch the replay of this insightful webinar, where industry experts shared valuable strategies for navigating uncertainty and leading through disruption. You’ll gain strategic insights from industry experts and learn practical techniques from CoPilot, our AI-powered guest speaker, on how to enhance decision-making, communication, and collaboration.
With a focus on leading through disruption, this webinar will inspire you to take action in driving success for your team. Don’t miss the chance to engage in an interactive Q&A with CoPilot to explore innovative approaches to leadership!
Why Attend?
- Strategic Insights: Gain strategic insights from industry experts and thought leaders. How do we use this technology with appropriate guard rails.
- CoPilot Mastery: Our special guest speaker, CoPilot, will share practical techniques for enhancing decision-making, communication, and team collaboration. CoPilot’s unique perspective, fueled by AI, will challenge your thinking and inspire action.
Agenda:
- Introduction: “Navigating Uncertainty: Leadership Lessons from CoPilot”
- Discussion: “Leading Through Disruption: Strategies for Success”
- Interactive Q&A: Your chance to engage directly with CoPilot
Watch the Webinar Replay
Transcript:CEO Mastery – Empowering Leadership with Copilot Collaboration
Introduction
Thank you. Perfect, all right, I want to welcome everyone. And today we’re going to be talking about a CEO topic, looking at co-pilot from a business perspective. And the reason that we’re doing this webinar is because at Certified Nets, we are starting to see our existing customer base investing CoPilot and use CoPilot successfully. And we’re starting to see some patterns develop in terms of their use cases, how they’re using CoPilot, how they’re implementing it, and what’s contributing to that success. We also have a guest speaker on here with us, Joshua, and we will be introducing ourselves. But If you do have any questions while we go through this information, if you could put it in chat. We will definitely be circling back and doing a question and answer and we’ll be reviewing those to see if there’s timely questions throughout the discussion as well.
Speaker Introductions
Robyn Howes Introduction
So, I am Robyn Howes, owner and founder of Certified Nets, and we’ve been in business for 23 years. Prior to working at Certified NETS, a managed IT company based in St. Louis, I worked at personal performance consultants and bridge information systems when they were small to medium-sized businesses. Both of those entities went through huge mergers and acquisitions and became international organizations, Merck for one of them, Reuters and Savvis for the other. And at the end of those great opportunities, I learned that I enjoy working with small to medium-sized business owners, being able to make decisions, being able to roll up our sleeves, being able to get things done. So over the last 23 years, Certified Nets has developed a unique process of how we can help our customers reduce the noise, so reduce reactive issues, so that we’re able to spend 70% of our resources helping our customers with more proactive and strategic items than just reactive. Our customers tell us that we’re a customer service firm that happens to do technology, and we’re an award-winning managed IT service and cybersecurity provider. We are lucky to have Joshua here, and he is a Microsoft Cloud architect.
Cloud Architect Joshua Hickok Introduction
Yeah, thanks for the introduction. Super happy to be here and want to just thank everybody for donating some of their time to listen to us talk about this. So I’m a cloud architect. I’ve been in the business since the 2000s, started as a Dell and Apple repair tech and kind of went from there started on the hardware side built my way over into the consulting side. Started as a security consultant, spent a lot of time doing that, moved into Cloud, became closely aligned with Microsoft. As Microsoft became a Co-pilot company, then all engineers that worked with Microsoft, all became Copilot engineers. That’s what I’ve been doing. I’ve been doing it for a few years now, really like it. I have the pleasure to serve on my company’s AI Center of Excellence as well. So that’s been kind of changing at a torrid pace. So really excited to talk to you guys and feel free, like Robin said, drop questions in the chat and we’ll get around to those towards the end.
Agenda Overview
Excellent. All right, so as I mentioned, the reason that we’re having this conversation today is we are starting to see a growing investment in Copilot, growing usage with it, as well as some best practices that are coming out of this. And we want to share that information with you. So in terms of an agenda today, we’re going to be talking about:
- Why is this a CEO topic as opposed to a technology department topic?
- We’re going to talk about if you do choose to embrace Copilot or artificial intelligence, what sort of guardrails do we need to have in place?
- We’re going to talk about the use cases that we’re seeing multiple customers implement successfully so we can share that information and
- We’ll culminate with some questions and answers.
Why is this a CEO topic and not a technology department topic?
Business Impact and Evolution
So first of all, why is this a CEO topic and not a technology department topic? What we are seeing is as people look to embrace CoPilot, we are seeing success in terms of innovation and creating efficiencies. We are seeing an increased investment in this technology and that this technology is definitely gaining in market share. So since we’re seeing benefit from it and in growing investment into it, it’s something that as CEOs, we need to pay attention to and we need to decide if the timing is right and the potential is right for us to embrace this technology.
Historical Context
So, historically, artificial intelligence was a tool that technical professionals wanted or our employees wanted, but from a company standpoint, we had to be careful about if it put our data at risk. So, oftentimes, at a corporate level, we would discourage the use of artificial intelligence. A lot of times, we would create policies that employees would need to sign because we were concerned about what data was being put out externally and who it was being shared with. We also have seen issues where either our clients viewed our customers as vendors and were concerned about their use of artificial intelligence with the data that those clients had access to. And so different compliance requirements would come from above about using artificial intelligence. But what we’re seeing now is a bit of a change in that tide.
Customer Perspectives
And so here are some quotes from customers that have chosen to embrace Copilot or embrace artificial intelligence. So we’re seeing customers say that they recognize that artificial intelligence is a powerful tool. They want their people to have access to this tool early and have the ability to experiment for themselves as well as for the company, how this tool could be used in beneficial ways. We’re also seeing customers who view this tool as an ability to automate.
Employee Impact
So looking at Excel, that for example, their controller processes in Excel that their controller would have to do on a regular basis or how to screen out emails. And basically within their organization, they viewed having a few small couple users with Copilot being able to experiment, find successes, and then train the rest of the staff is how they like to incorporate technologies such as this.
So going back to why is this a CEO topic? Basically, this is an example of where in uncertain times, this is a potentially disruptive technology. And so it lends itself to being a time to experiment as we’ve heard in those quotes above.
If we think about Patrick Lencioni in the incredible book, Great by Choice, where he talks about fire bullets first, then cannonballs, this is a great example of how we can make a small investment, designate a couple of users to begin to experiment with this technology, and then ultimately it may have the potential to give you an edge in the marketplace. place.
In addition to the business topics of why this is a CEO topic, it also impacts employee morale. So what we find is that by investing in artificial intelligence or Copilot, employees view this as an investment in innovation and an investment in their learning. And so they view this as cool and positive in terms of what their company’s goals are for the future? And Joshua, I know this is a topic that’s top of mind for you.
AI Empowering Employees and Streamlining Workflows
Joshua: Yeah, it is. Companies that have what I’ll just call a progressive position on AI are lucrative to employees. I actually saw my first job posting on LinkedIn. I wasn’t looking for myself, but I saw one out there, somebody drew my attention to it, where they actually pointed out that one of the benefits working for them was they had invested in AI tools. This was for a BI position, a business intelligence position, where they would have access to some of these. Even job postings now are putting it out there as a draw.
But one of the reasons why it’s becoming important for an employer to at least have a stance one way or the other, is that a lot of employees are increasingly using these tools on their own, as you mentioned. And AI, what we’re seeing is that it has the ability to take care of, it can help you with high value work for sure, but one of the biggest things is it can very quickly kind of remove a lot of perceived low value work or simple work. And so it’s a layer of automation, but it’s not that kind of unseen black box automation. It’s that sort of thing that lets you work with it, right? It’s truly an assistant. And so you can kind of fine tune it and get what you need out of it.
So we’re seeing employees now, potential employees, starting to expect that you either have these available or you’ll let them use their own. And I love this last point that you have up here about the CEO decision, right? The CEO decision is either an active decision that you’ve made or a passive one. You either have a stance on using AI in the workplace or you don’t. And not having a stance is maybe the worst position. It means that you’re just opting out of governance. It reminds me of some of the places, some of those businesses that took a long time to build a work from home policy, right? And so by not really addressing it, the workforce is going to build their own kind of tacit policies. So it certainly can make sense to investigate. In this case, we’re talking about Copilot, but AI in general and say, it doesn’t seem like it’s ready, it doesn’t fit our use cases right now. That’s okay. What we want to avoid is just ignoring it and kind of refusing to weigh in on it, either actively saying, I don’t want to think about it or passively by just delaying. So I would, even if, you know, kind of today, it doesn’t look like it’s time, keep checking in. And because the market has spoken, right? Employees have spoken. The demand is there, it is coming. The question is, do we have those use cases today to make kind of that mature, good response.
Guardrails for Artificial Intelligence Tools
Robyn: Excellent. OK. So if the response is that we want to embrace Copilot, what is our risk? What sort of guardrails do we need to keep in mind? So what do people see internally? Is the information shared externally? And what does that risk profile look like compared to what it used to look like with CHAT GPT?
Security Model
Joshua: Yeah, so that’s brings us to kind of kind of hear what I want to talk about is we have four main points that we wanted to talk about today. So one, the model of security with Copilot is modeled after the same stance that they have for M365. There’s always risk when you use a piece of technology, right? It has been since the beginning using a laptop, using the internet, these things invite risk. The nice thing is, is this is the same risk profile as using M365, which is pretty much universally adopted at this point. To break that down, so your data is always yours. You can delete, remove it at any point in time. Your data is stored in your tenant. This is the logical boundary of your organization. This is where your Copilot’s data is stored, right? They give it the fancy name of a semantic index. That’s just a corporate database that these large language models use that’s stored with your emails, with your chat, that sort of thing.
Data Protection Features
So that is the security model that if you’re using M365, you already kind of approve of. Your data at rest and in transit, again, uses the same technologies. These are proven time and time again, it’s good enough for the US government. Then they defend your data because they’re running a private model. OpenAI, Microsoft isn’t sending your data to OpenAI for processing. Microsoft has access to OpenAI’s latest and greatest models. They’re able to run those in a private secure instance inside of Azure, and this is what you’re working with. Your data stays with you, interacts with Microsoft’s private models, and is never sold. Your data is never training a model. It is completely and logically separated from everybody else’s data. So no worries there. This is a model that really is unsurpassed in the marketplace. Almost no other company has the ability to follow those logical safeguards because they’re also serving you your data via email and chat. And so this isn’t an add -on, also a must -secure tool. They’re able to slot this in if you’re already a Microsoft customer today. I think we can kind of move to the next one.
Access to Powerful AI Models
Yeah. So just quickly here, one of the best things is, Robin, I remember when I first used GPT, there’s a magic to it. That magic was, there’s no way this is going to be any good. I’ve worked with chatbots before. You do it all the time when you’re trying to return an e -commerce item or you’re trying to get through to support, and you’re trying to figure out a way to hack the bot so that you can get to a real human, that sort of thing. ChatGPT was different. When that first came out, I was like, wow, this is great. They’ve increased the technical proficiency of those models, and it’s just been a leader in this space now for a few years.
The good news is Microsoft CoPilot is just built on the latest and greatest from OpenAI. OpenAI has partnered with Microsoft and Microsoft has exclusive access to these. Your point there is really good, the additional cost of Entry ID. Your identities, if you’re using M365 today or even if you’re not, you can use your identity to interact with them. The hope is that you’re securing these identities already, and the great thing is you can reuse your current security posture to interact with these models. It’s not, again, some additional tool. The transparency and currency is super important.
Transparent and Current
So one of the things that we require, which you might not get from ChatGPT, is the sort of citation that you require in order to be sure of your answer or to protect against the problem of what they call hallucinations. And so, for example, if I were to ask Copilot Chat, which I think you’re going to get into and I don’t want to steal your thunder, Robin, is I could ask it a question about, I’ll raise my hand.
My company is actually building Copilot into our e -commerce platform. I wasn’t sure when that date was when we’re actually releasing at GA. I was able to pop open Copilot chat and I was able to ask it, hey, what date is this being released? Now, it could have told me a date and I could have just believed it. But what was important is it actually went and it found the transcript of a meeting where the President of America was talking about this and announced the release date. I was able to see that in the transcript. Then I could go to it, I could click on it, I could go to it, and I could hear it live for myself. That’s really important when it comes to the business world. ChatGPT is fantastic for individual use, I have no issues with it, but you need something a little better, a little more concrete with some of these additional features. so the ability to cite corporate docs is super important.
Commercial Data Protection
Then moving along to the commercial data protection. Again, you have these tools already if you’re a Microsoft customer to secure how your large language model activity works. Your chat data is private to you, so you don’t have to feel terrible about asking it something silly like, Hey, who is my new boss? I moved to a new department, right? That’s not gonna be sent to HR. Like, hey, this guy doesn’t know who’s boss is. It’s a silly example, but the point is, is that’s a private interaction. No eyes on access, that’s great. Nobody can kind of, IT, Microsoft, they’re not able to pick you out and follow Josh Hickok’s activity, what they’re doing inside of there. And then, again, we mentioned this before, but your chat data is not used to train the model. Just categorically, it’s not. Some people are super concerned about that, and they’re never going to be super happy with the fact that they’re interacting with an AI. And there can be good reasons for that, but those aren’t technical reasons at this point, right? It’s a private model. And actually, training a model is a big deal also, right? It takes months, and at this point, we’re encroaching on billions of dollars to train. Your corporate data is not getting ingested for training. I think that’s what we want to say here, and I think we can move on to some use cases, Robyn, unless you have anything to add.
Use Cases
Robyn: No, that sounds great. All right. For use cases, what we’re going to do is we’re going to really focus on almost two different ends of the spectrum in terms of what we’re currently seeing amongst our customers.
Power BI
One of those use cases has to do with Power BI. What we’re seeing is where customers are using their financial information or their databases, and they’re using Copilot for Power BI to basically slice and dice that information, whether it’s dashboarding, reporting, but making that information much more readily available. We’re also seeing Copilot for Power BI used with regard to coding, making it easier to code requiring less skill.
Microsoft Teams
Then the second use case we’re going to focus on has to do more with Microsoft Teams. It’s got to do with our ability to use Copilot for collaboration and really some cool stuff with regard to meeting automation.
Joshua: Yeah, definitely. And I think on this, I just wanted to touch on a few things if I could Robin. So first of all, to do some kind of disambiguation, Power BI has recently gotten a facelift and I would ask for a show of hands who actually who’s familiar with Power BI. But Power BI is a business intelligence tool that sits on top of your data and gives you useful visuals essentially. At a glance, what does the data mean? Power BI has recently pushed down the stack, means it’s taken on more of the data capabilities. This is what we see, this first flag in the top. It now has the ability to manage data warehouses, your Microsoft assets, and even external assets. So it’s in kind of the space if you’ve heard of things like Snowflake or Databricks, some of these other ones. Microsoft has extended Power BI, kind of given it this name, Microsoft Fabric. The really cool thing is like you mentioned, Robin, it’s got some fantastic features that are AI -driven in a really concrete way. I think we’re going to look at that in just a sec, but the two that I want to point out are coding and natural language, so that’s big.
Data engineering now uses at least four different programming languages. I don’t know all four of them. I’m never going to learn all four of them, and if I do, it’s going to be around for a couple of weeks until I move on to something else, and then I have to brush off the cobwebs again. But you can now be, I remember seeing the quote that the next great programming language is going to be English, and the point there is that if you know what you want to see and you’re able to kind of type that pretty well, you’re gonna be able to code in natural language. We’re gonna see that. And the AI created data dashboards. And so I think we kind of added a little visual here if we want, Robin.
Power BI inside of Fabric
So this is Power BI inside of Fabric, and you can see that you can interact with it in natural language kind of on the right -hand side, and that’s usually where we see CoPilot, right? But you can ask it very kind of highh-level questions or very specific, and you can see that what CoPilot is doing is actually driving the creation of your, you know, your KPIs, your visuals, but you can fix it. If there’s something that you don’t like, you can fix it in natural language. Hey, it looks like CoPilot, you did this, I actually want this a little bit different. Or could you turn this from a bar chart to a scatterplot or something like that? And you can continuously interact with it till you have it right.
Coding in Natural Language
Now, this last part, this is the coding in natural language. What we’re seeing on the left is DAX. Now, I’m not going to quiz you guys on DAX, but I suspect that most of you don’t really know what DAX is. The nice thing is that CoPilot will build this for you, help you iterate over it, and get the results down at the bottom without ever knowing what DAX is. I’ll raise my hand on that one. I am like an F-tier DAX report creator, which is something that’s required to create the most advanced visualization and analytics inside of Power BI. But it’s a Power BI specific language. You asked me, I think, Robin, about visual basic but really macros, right?
And so if this is something that you would like to see, you can actually transition from these kind of older technologies that Microsoft suggests you move away from without learning its replacement, which in this case is DAX. So very powerful. You can see how it works. And understand now how, I hope this is supposed to be kind of a visual of how Copilot can differ from ChatGPT, right? It’s not just private, secure access to a large language model. It has integrations into the tools, right? Otherwise, I’m taking screenshots of my corporate data and pasting them into ChatGPT, probably some red flags there anyway. but the point is ChatGPT cannot interact with Fabric slash Power BI. I think the next example that we’re going to get to will show that.
But if we could just move on to this one real quick. Fabric, again, just to disambiguate, is what happens if you want to take Power BI to the next level. This is where we see that Copilot for Power BI come in. We’re happy to take some additional questions on that. if you’re interested, but the point is this is a data lake. This is a flat storage place where all you have to do is get your data into it, and then you can let AI do the rest, at least a lot of it. It still is an assistant, you still have to prompt it, but you can build those automations for your data engineering, for your pipelines, that sort of thing, From start, from the warehousing, which is usually how it starts, let’s get the data in and warehoused, all the way to the end where I’m publishing rich visuals for the rest of my organization to consume. I’m doing all of those steps, not by learning Python and Scala and R and DAX, but by just learning how to prompt well. That’s an iterative process too. It’s something that you learn and you get used to. It’s a really interesting platform. It’s a really interesting use case of infused AI into a Microsoft product, which I think then takes us to probably my favorite, Robin, which is what you’re going to talk about.
Collaboration & Meeting Automation
Robyn: The other use case that we’re going to highlight has to do with using Copilot and more of a collaboration way and meeting automation. So when we talk about this use of Copilot, we’re really talking about Copilot for Microsoft 365. This is, it’s an annual commitment, so it’s $360 per year per user, but this will provide the functionality that we’re about to review.
So for example, with Copilot, we can put in a keyword and it will search our Microsoft assets. In this example, it’s looking at two Word documents. We’re asking it to compare and contrast. We can then also ask Copilot to instead, using these two Word documents as content, turn this into a webpage format for us. So it’s saving us from the work of doing the compare and contrast, as well as combining that information and turning it into a different format. So the bottom line is, when we think about Copilot, it has the ability to become our search engine, much like we’re used to using Google for websites. Copilot can look through all of our Word documents, Excel files, PowerPoint, SharePoint, email, chat sessions, basically everything as we give it our keywords that we want to research. So it puts less and less pressure on us having a good file structure and more emphasis on our ability to ask it good questions.
So as we create those good questions, we want to think about our goal in this example for it to create something with three to five bullet points, or we could say two to three paragraphs. We often wanna give it some context, like this is gonna be for a meeting with client X. We wanna give it some idea of sources to use. So in this example, please look at my emails and my team chats during the month of June and give it some expectations of its output. So we’d like it to be in plain language so that we can get up to speed quickly. So that’s just an example because the more we can provide good queries, the better it’s gonna be in terms of giving us the results that we’re seeking.
So in addition to using CoPilot as a search engine, it’s very, very powerful with meetings. So if we think about a time where we’ve had two meetings and we want to attend both, but obviously we can’t, CoPilot gives us the ability to follow one of the two meetings. So for example, if we, and this is a brand new feature that just came out in June, but we can choose to follow one of the two meetings. And what will happen is when that meeting is completed, it will send us a notification. And when we click on the notification, it provides us a summary of the meeting that we missed, as well as outlining the tasks that came out of that meeting and who was assigned to each of those tasks. And then we can even ask CoPilot some clarifying questions. So we could say, what other customers were discussed? Or why was the decision made that was made? Or were some alternate decisions thought about? And we can even click on some of those results and get down to an actual transcript from the meeting itself. So very granular, very powerful.
We can also use Copilot during a live meeting, not just from a recap standpoint. We can ask it things like, can you summarize the meeting up to date as of now? We can ask it, can you give us a sense of how the group feels about a particular topic within the meeting? What other questions remain in this meeting that are unresolved? So it’s pretty powerful in making sure that the meetings that we are attending are accomplishing what we need them to accomplish. So in summation, Copilot can help us during a meeting in a variety of ways, and it can help us after the meeting from a recap standpoint. And what we’re seeing from our customers is this recap component is what we’re seeing them use successfully. So, the ability to summarize, the ability to create these follow -up tasks, sometimes these summaries go into their CRM systems, so they have a much more robust set of notes, more streamlined, and then in terms of internal processes, really taking advantage of these follow-up tasks to make sure every resources are tied to these tasks and they are completed.
So, so far, we have talked about Copilot for Power BI. We’ve talked about Copilot for Microsoft 365. But the truth is, there’s other versions of Copilot as well and even more coming around the corner. So Joshua, if you want to speak to those.
Joshua: Yeah. So what we’ve talked about are, we talked about a specific use case. we talked about Power BI, which is AI infused business intelligence. The Copilot for M365 is a generalized toolkit for across the entire M365 family. It’s included in Excel and Outlook. We’ve talked about a couple of those. But there are these point solutions that are intended to address a persona or a specific use case. So for example, there’s a number inside of Dynamics. So there are Copilots that can help you that integrate your email, your chat, and your CRM, right? So it does support Dynamics, it also supports Salesforce too. And so it kind of brings those things together. And just the way that it can look is, you’ve got a meeting on your calendar with a potential client.
You call, you talk to them, CoPilot understands your conversation, puts the details into your CRM for you. These are the auto attach notes. The idea is to increase the velocity of how much you can get done by removing that busy work for you, for things that an AI can do really well. I also want to just point out CoPilot for Security too. This is their most recently released paid CoPilot, which is like adding a new SOC analyst, but one that has a full 360 view of your organization along with your policies, along with events, and so you don’t have to go and search for these things and bring it together on your own. Really, really powerful. If you have a security operations center, it’s a cool addition there. The idea is for all of these Copilots to automate as much of that busy work as you can to get your high paid valuable employees doing that valuable work.
Robyn: Excellent, all right. So as we conclude, I would encourage people particularly those less familiar with Certified Nets to email info at certified-nets .com with your top two tech challenges. And that’s an opportunity for us to talk about next steps and talk about what it is like to work with a partner that is focused on some of these topics that are more strategic in nature, while always keeping productivity and security in mind. And next, we are gonna address questions and answers. So I see some people have written into chat, so we’re gonna talk about those. And we’re gonna start off with just a couple of questions that customers have asked as we’ve been discussing this topic with each of our customers individually.
Q & A
Q: What is the difference between the free version of CoPilot that we all see, usually within our taskbar, versus some of the paid ones like Microsoft 365?
So one question we often get is, what is the difference between the free version of CoPilot that we all see, usually within our taskbar, versus some of the paid ones like Microsoft 365? So Joshua, if you want to address that.
Joshua: Yeah, and that’s a great question. So as you may know, A copilot is a, it’s kind of like a family, right? So it’s a family of items that you can use. Some of them you can use at no cost, right? So one of them is copilot or the artist previously known as Bing Enterprise Chat, which you can use. And this is a way to use kind of like a safe sandboxed version of ChatGPT. There’s a lot of value in it, for sure. The difference between that and paid is what they call the semantic index, right? This is what I mentioned is your database built of your corporate identity, your files, your folders. So when you pay, Microsoft builds out this database inside of your tenant. And so what that looks like to you as an end user is you now can use this large language model and it now is aware of your corporate data. I can ask it questions about emails, I can ask questions about chat, I can ask it about documents, all of those sorts of items. And so that’s the difference, is you get that corporate database that then you can kind of work with and it does what it’s called it’s grounding of its answers through that corporate database. So it gives you that contextual information. It’s a good question. Excellent.
Q: If we know we’re going to start to use artificial intelligence, do we need to look at our computer hardware and maybe invest in our hardware differently?
And then we’ve also been asked if our company, if we know we’re going to start to use artificial intelligence, do we need to look at our computer hardware and maybe invest in our hardware differently? like making sure it’s got graphics cards.
Joshua: Yeah, so the the great part is that Microsoft you do need NVIDIA hardware right? But Microsoft did that for you. So these models that are hosted in Azure run in Azure when they’re when they’re coming up with their answers to your prompt. So Microsoft purchased all that NVIDIA hardware runs them in their data center centers that is required, but it’s not required for you to invest anything right? So it’s a SaaS offering. All you need is a computer with Internet access and then you’re good to go. Good question.
Robyn: Okay. Excellent. I know we talked about macros earlier. So I would just keep in mind that macros is an example of a technology that is more legacy in nature and between Power BI and Copilot for BI, there are paths of no longer using macros. Joshua, I don’t know if you want to address that quickly or we can go to the chat.
Joshua: Yeah, just real quick. If you’re using macros, God bless, those are a security vulnerability at this point. The whole point of DAX is to be a VBA -like language, so you can get the same result but in a modern tool, and a modern tool that’s easy to share without sending something insecure in a file over the Internet to your friends who are going to get warned. This file contains macros, we’ve disabled them by default, are you sure? That’s not a message you want tied to your name. There’s a great replacement story there with Power BI.
Robyn: Excellent. Okay, so I am going to or we are going to look at the chat. So some of the questions that have come up.
Q: Do we need to worry about copyright infringement or crediting AI for generated content?
Joshua: Yeah, so I can so. First of all, I’m not a lawyer, so take everything with a grain of salt. This is not legal advice, so there is a lot of interesting work that needs to be done in this space, right? I don’t want to kind of go off on a tangent, but some of this is getting worked out.
You typically what we’re seeing is that. You are not liable, again, not a lawyer, but users of it are not liable. But in fact, to the creators of these large language models maybe. So we’ve seen some payouts and that sort of thing. But at this point, there’s no reason to think that you’re somehow legally culpable for using large language models. So again, not a lawyer, but this is what we see. We see some people just started to take payoffs. So Reddit, which was one of the early sources that OpenAI used, started gating access to their data with an API that OpenAI happily paid, so you’re okay to use it. So I think that’s the sort of thing, little speculation on my part, but I think we’re just going to start seeing payouts to some of these content creators. Do you want to move to the next one here?
Robyn: Yeah.
Joshua: So mention of sighted source transfers, how are these sources filtered? So as a SaaS product, you don’t have a ton of say in kind of how things are presented to you in a proactive manner.
The idea though is that these things are conversations, right? So part of this is going to just fall to prompt engineering. So this happens all the time, right? So I use kind of enterprise search all the time, right? Copilot .Microsoft.com. That’s replaced my SharePoint search. I don’t search through. We’ve got 22 ,000 employees, all sorts of data out there. it’s impossible to do keyword search. I rely on Copilot to find things all the time. Sometimes it’s a back and forth like, hey, I see what you’re getting after Copilot, not quite right. This is actually what I’m looking for. You can make suggestions like I’m actually looking for a PowerPoint. Send me the PowerPoints that are about this. That’s how that works. Reporting an incorrect source, you can thumbs down and that can tweak your semantic index a little bit, it can make some changes there. But usually, again, that’s a part of the prompt engineering. It’s a good question. For a SaaS offering where a lot of that is, quote -unquote, black boxy on how it works, you don’t necessarily have a lot of power over that.
I will just add that there are, if you really do need a personalized bot, there is a thing that you can do. You can get Copilot Studio and you can build your own Copilot. It’s actually kind of easy, but if you find yourself needing something a little more custom, that is an option for you. Excellent.
Robyn: All right, and for that third question?
Joshua: Yeah. So the third question says, while using Copilot, there are certain things when generating content Copilot is going to ignore. Are there ways to emphasize specific aspects to get the results you need? I run into this a lot while using Dolly. Good question. I think the answer there, if I understand the question, is going to be similar to two, which is prompt engineering. Prompt engineering is more annoying when you’re dealing with media creation because the process is slower. You know it’s it takes longer to generate kind of an image than it does a sentence. So, but the answer there is probably prompt engineering. There’s also other things that you can do. So if you have items that are showing up that you don’t ever really want to see, you can blacklist items, right? So we have SharePoints that are noisy at my company that we didn’t want appearing or because they were, you know, it was like HR related and we blacklisted entire areas of our org from a data perspective and Copilot just doesn’t index. It doesn’t see it, can’t respond from it. The other thing too is if you have a specific use case, I again would maybe investigate Copilot Studio because you can build in prompts that everybody will then see. So you don’t have to type it every time. So you can say, hey, you’re a support agent, act as a support agent or hey, make sure that when you’re building these images for me. Which you know that’s a really personalized bot, but you can if your org wants to use it, you can build that sort of thing with those prompts built in and give it a certain sort of behavior that will persist then through different sessions. That’s a that’s a great question.
Robyn: OK, and then the one last question in terms of members of the hosting org being the only ones to have access to Copilot. Do you want to address that, Joshua, in terms of inviting external users?
Joshua: Yeah. It’s true. The reason is because this corporate database uses your user permissions, your inbox, your chat, and it’s not going to build a semantic index with external users. For one, because you can’t just read other tenants. I can’t just, Robin, I can’t go into your tenant and understand what permissions you as a user have, that sort of thing. There are workarounds for sure. So I can add an external user to my tenant and give them a persona inside of my tenant, and then I can add their activities here to our semantic index, but it’s never going to just read an external tenant. With that being said, my understanding is that there’s going to be a way in the future to whitelist a neighboring tenant. There are some companies out there that have more than one tenant, and so that hosting company, you will be able to form relationships with other quote -unquote hosting companies, which to Microsoft look like external tenants, and then you will be able to, is my understanding coming soon.
Conclusion
Robyn: Excellent. All right, so with that, I want to thank everybody for their time. And again, this is one way that Certified Nets is hoping to add value in terms of education and advising and collaboration and strategizing. So I hope you found this helpful, and you all know how to reach me. So if you have additional questions or technology challenges, we would love to help. Thank you for inviting me, Robyn. I appreciate it. Thank you, everybody, for attending. Thanks for joining us. Have a great day, everyone. Bye-bye.