Transcript
So, guys, today's topic is how to turn the AI insights into IT data and vice versa. How to combine what you're doing today as an IT practitioner, regardless of whether you belong to the IT department or to a managed service provider. How can you put it to good use? How can you optimize the time you spend on the ticket? Maybe something to do with your regular day-to-day admin tasks, which no one likes to do. Muna, maybe you can share anything you heard from customers recently. No, obviously, again, I don't think anybody... I want to assume that everybody here is using AI in one form or the other. Obviously, it's heading a specific department. One of our key strategies is to make sure AI in one form or another is being used day-to-day. And we all look for ways to simplify, maybe add productivity, you know, sometimes just to get some ideas of what AI can put forward in different domains. And if we're already talking about AI, and because we're here to talk specifically about Dima, I'd love to engage the audience and see who do we have in the audience. So please, I'm going to share a poll. You will see a poll button to the right of the Q&A. I'm going to open it now. Please go ahead. Let us know. Are you currently using Atera's AI Copilot? Is that a yes, a no, or here to explore? Again, we're here to focus more in-depth on some additional use cases, but we'd love to know. So please don't shy, let us know. Dima, I see that the majority here are between exploring and interested. Wonderful. Amazing. Then we have the very right audience for us. So without further ado, while everyone is finishing with the poll, let's start the actual webinar. And because I've seen one of the questions in the chat from Michael, like, what's AI? So I think that's actually worth a little bit of explanation. I won't go into too many details, as I always do. I just love details. Don't worry, Dima. We'll set you back into the demo, okay? Yeah. Okay. Thank you, Muna. Okay. So AI in general is something that probably it's very hard not to notice in 2025. Many cool companies releasing cool products. But basically artificial intelligence is something that came after the machine learning to revolutionize how it works. And right after the traditional AI came the agentic AI, which is something that is very new to the industry. And it's basically something that is not just a chatbot, which can answer any random question you throw at it. It has a certain knowledge behind its back. But also with the agentic nature, if it has access to your device or to a certain system, it knows some additional context. It allows it to be smarter. Anything that I missed, Muna? Was it short enough? No, that's good. I think some of our audience is keen to differentiate between what is AI Copilot and the other AI offerings. So I'm sure that you'll be talking about that in just a minute. Yeah, exactly. So here's the agenda for today. We did a little bit of introduction, but I will just remind who we are in just a second because somebody might have just joined. The problem that actually is a burden to many of us, me included, that just what takes our time without us knowing. Something that I wanted to share with you guys. And maybe share some thoughts. Maybe you have some questions. Maybe you can attest to it. So put pluses in the chat. Then we'll go to the demo and we'll see how it works live in action. And I will connect the dots in between your day-to-day tasks and how you can scale and make the picture bigger and report to your management, report to your colleagues, your customers, and make more AI actions worth your time and make your IT day-to-day activities more meaningful by just doing less things you don't like. This is what our AI products are here for. But before we do, it's two of us today in the room. It's me and Muna. I'm Dima. I'm the Technical Product Marketing Manager at Atera. And Muna, can you please introduce yourself? Yes. Hello, everyone, and welcome. I am VP Product Marketing here at Atera, and I'm very excited to be co-hosting this webinar with Dima. I will say we run these webinars on a biweekly basis covering several topics within Atera's core solution as well as our Atera IT autopilot. So please make sure to register to these webinars. Both of us are here and more than happy to help you make the most of your Atera solution. Thank you, Dima. Okay. Thank you. Okay. So we already had this question in the chat, and I want to provide you with like a short explanation, but I'll try to make it as meaningful as possible. So in short, Atera has two AI products, one of which is focused around your day-to-day activities as an IT practitioner. So it's called AI Copilot. So it helps you to do your day-to-day IT tasks while being in the Atera admin console. Second AI product that we have in our portfolio is IT Autopilot. Autopilot is somewhat focused around technicians, but the main audience is apparently end users. It's something that can complement your customer portal where users submit their tickets, and basically Autopilot does everything a tier one technician would do, do initial troubleshooting, answer basic questions, maybe not IT related. And so it kind of stands as a gatekeeper before you get to do the ticket, which requires some actual human attention. And the Copilot complements what Autopilot did, helping you to respond to your tickets, summarizing them, generating KB articles, helping you to level up your ticket automations, which we'll talk about in a second. And anything that is in the realm of the ticketing, but also Copilot does an amazing job in the RMM realm, helping you to diagnose devices without actually accessing them remotely, just saving time by not asking your end users, is it okay if I do the remote session with you right now, stuff like this. And Copilot doesn't end up there. So basically Copilot and Autopilot work together. They're both fueled by your knowledge base articles, they're both fueled by your internal company knowledge, so they are not only like general IT knowledge, but also your company specific knowledge. And this is where our AI product starts to shine, because how would it know that it's you talking from your computer, and there's no way around it. Muna, anything from you to add on the topic? No, looking forward to hearing more about the use cases. And if we're already talking about it, and definitely Dima, you're going to go into the details. I just want to bring the attention of the audience to the button at the top right where it says an AI Copilot playbook. So today Dima will focus on a few very, very useful use cases. Dima spends a lot of his time training clients, onboarding, doing in depth professional services. So I think he's seen all of the different use cases and best examples of how a lot of Atera clients are leveraging Copilot. So he's not going to share all the use cases today. So we did add this playbook that will give you ideas of how best to utilize Copilot. That is true. So, yeah, I've spent a lot of time, both in customer success and professional services with tech focus. And I did the most complex Atera implementation across the customer base. And yeah, I've seen most of the use cases, not all of them, but I wish I did, but I didn't. Anyway, before we go into specific use cases, I want to start simple. I want to talk about something very important, and I haven't seen much about it out there. I didn't see that many LinkedIn posts or something like this or maybe credit threads. But how do we interact with our computers? So as any user on planet Earth, we use the keyboard and mouse. Some people use the voice input. Some people use any type of alternative things to type in text. But we all do type text because this is the prime interaction method with a computer. So, and I want to start with a fun fact. So when we're composing something and we need to think, we type at 20 words per minute. You might be thinking, why am I even sharing this? But bear with me. It's going to get interesting. Trust me. So when we're typing something that we're very well aware of, we type in between like 40 to 60 words per minute. Some advanced typers, they type above this level. I don't know the exact numbers, but it's close to like 70, 80. And with certain type of keyboards, it may be even above it. So what does it have to do with your day-to-day work? Well, it takes your time. So you're limited with how many words per minute you can type. You might know the answer. It might be sitting inside your head, but you don't know how to express it. And this is where AI tools come in because apparently they type and this is the amount, this is the value that you see, 350 words per minute. This is how the UI is set. Why is it set to like approximately this value? It's because in the background, it types several like dozens of times faster. However, for your convenience, when you try to read the thought process is about 350 words per minute, depending on the model and the product. It is done for you to, so it's easy for you to follow its chain of thought and catch it if it drifts away too far. So, and apparently I've come across a study made by the Zurich University of Applied Sciences. Yes, it's from 2021, but apparently there's not much study on the topic. And they started more than 1 million tickets across many industries. And they came to a conclusion that the average IT ticket word count is 31. Which is, which kind of give you an example of like what would take you one minute to respond will take AI significantly less to do exactly the same. I mean, to write a response for you, suggest a response or actually read the content and understand it. And where does it come into play? And as an RMM and PSA products users, you might be interacting with this specific areas. So to name a few, it's like when you're responding to the tickets, when you're trying to look into the alerts, you're trying to understand what it is. Maybe you want to leave a reply to a ticket. Maybe you want to leave a note. Maybe you are as an IT technician, you're escalating a ticket and you need to supply your colleague with an internal note. What does he need to do? After every ticket, most of the customers that I met, they need to do one very exciting thing, which everybody likes, of course. Yes, it's irony. And the resolution notes, because every time you close the ticket, most of the companies force you to do it. So to keep track of the knowledge of how to troubleshoot, maybe the new employee comes in for future use for any type of data indexing purposes you might have. KB article composition. So the resolution notes for successful result tickets might lead you to creating a KB article on the matter. To actually put it to good use and reuse it for maybe further troubleshooting, training a new employee, and maybe giving some more context to your colleagues. Because KB articles can be just for the IT team and they can be for the end users. And it actually works well for everyone. So the escalation pack I mentioned, and the last but not least, but extremely important because it all ends up in reporting. How do you explain it to your manager? Why do you spend 16 hours on a ticket but not one? Maybe you needed one minute, but maybe it ended up with several days. How to justify the complexity? What was the ticket about? What steps did you take to troubleshoot? What was the actual path from the very beginning till the very end? What made you succeed with a very specific IT case? So why am I mentioning these two things together? Well, our co-pilot is touching all of those. And you might have a lot of extra custom fields, information fields, anything of that nature that needs your input. that needs your input. And we get back to our typing speed, OK? And let's move forward. And let's study the impact that AI can do for a company which is pretty small and is not doing that many tickets, maybe. Maybe for some it's a lot. But let's imagine there is a company with five technicians. They do about 200 tickets per month. We're still taking into consideration that we have the 31 words per ticket on average. And let's imagine there is a hypothetical hourly rate for a technician, which is around $60, sorry. So we all have a limited amount of time in a given month. And we need to resolve things. And all of those tickets that come in, they might require your fast response. Maybe you got distracted. Maybe you lost a chain of thought. Maybe you just started your shift and someone is just giving you the ticket and you don't want to read through pages and pages worth of text. Maybe you just need a quick summary to save your time and to get to do the actual work. With that being said, so if AI is applied to all, like in most of those areas, because according to data, according to Zurich University of Applied Sciences and another study made by the company called TZS in 2023, so customers of this size, like five techs, 200 tickets per month, they have approximately 30% repetitive ticket rate. Most of those tickets could have been a KB, which could have been easily accessed by the end user. And with that being said, a lot of users can be just fine with like a printed piece of knowledge without actually taking the IT practitioner's time. And with all that being said, so if AI components can save a company of this size, this amount of time, it can also mean that it can save money because your time is money and our hourly rate multiplied by hours results in a very humble number. And I don't think that like just typing, typing in all those places that I mentioned, like the ticket alerts, the ticket replies, all of it, all of what I mentioned, I don't think everybody like just, everybody just takes it as a part of the package, as part of their professional life. We don't treat it as our like actual effort. But to do the job is sometimes like to resolve a ticket is five times, but to explain it to someone who is not aware, who is not technical, who doesn't understand what you're doing, maybe they don't understand how the product is called, this is where the complexity comes in. So you need to put effort and go all the way down to 20 words per minute and burn your precious hours with your precious keyboard, probably mechanical one, that sounds nice. Anyway, so, and if you project this amount of like money that just 40 hours per month can save a team of this size, you can do an easy math to extrapolate this data to a company of your size. You can be a little bit less, a little bit more, depending on your workload, depending on the amount of technician, depending on the hourly rate. But the rule of thumb is somewhere around here. And Mona, I want to ask you a question. So did you know that it actually is that important? Do you feel like you're typing when the answer is right inside your head is just so, I don't know, it's haunting you to write that response, you know how to respond it, but then someone comes to your desk, maybe me, and it just takes your. And this is extremely valuable, Dima. And I think, you know, as you're going to go through this, one element of it is definitely the cost, which is a critical element, time, but also the element of the repetition, right? Doing more of the same over and over when some of that can really be helped through AI. I think that is another key area. Yeah, that is absolutely true. So now with all that information in mind, with all those data entry points, I would like us to transition onto the platform. And I would like us to see my screen. Let's do it real quick. I have, apparently I do have too many screens to choose from, but I'll try to do my best. So, okay. Please let me know if you can see it. And of course I will zoom in because to some. Yeah, we will see the platform first because this is exactly where all things begin. Just type in a plus sign or just put a like thumbs up like under the last Muna's comment. Okay, so I consider this as a yes. By the way, for anyone who is not seeing it at 100%, maybe it's too pixelated, at the bottom right or the screen sharing, there's a magnification tool so you can easily use it. Okay, so we just discussed many data entry points. Let's begin with the setup. So we have many cool copilot features. If you're new to the platform, let me give you a quick tour. So in the ticket realm, you may see the AI icon is located right next to any ticket that you open. And it basically gives you a summary, which is device aware, which allows you to get a summary and also ask device related questions. This is one use case for the copilot. So the second use case lives in the alerts realm because this is something that touches the RMM. So, okay, I need to summarize the alert for the computer and apparently it gives me an alert because I surpassed the threshold without even actually unfolding it. Another use case for the copilot, how you can elevate your day-to-day work is to start a general IT, like general IT, like start with a general IT question to make copilot understand what you're doing. However, to prove the point that Atera's copilot feeds off of your knowledge, you can just ask a question which might end up in a knowledge-based article because knowledge is key. So I'm going on a business trip. What do I do? I know it's a stressful question to some, me included. It may be nothing to do with IT, but sometimes it is because it might involve software access. It might require like a VPN settings and it might end up in an IT issue to some people. And they're already stressed. And as you can see, copilot picked up a KB article which gives you a step-by-step instruction. And it might've been, and your employee, which would ask us the very, very same question. And it might've ended up with you sending the same KB article, but you also could have been typing it in. So as you can see, the Business Travel Almanac covers many, many things here. I like, what is my daily allowance as an employee? Maybe like you want to ask this question. And yet again, it will have your company's knowledge and it will be absolutely fine with responding. So again, this is where the time saving starts. And as you can see, you have like your daily loans and also, I don't know, I'm going to, I don't know, I'm going to France. Can you convert my daily allowance? It can also be like a question which is somehow ended up in an IT ticket. So, and the copilot will try its best to give you the most recent one. So it still has the value in dollars. However, in certain, like I have knowledge there in the knowledge-based article, I wanted to prove something else, that it's not yet looking on the outside. So it lives in your company's realm. And this is why it's critical to put your knowledge inside its brain, because the general IT knowledge is there, but your company-specific knowledge is not. Okay, it has nowhere to get it from just yet. So how can I help you? So it can be just an easy copy paste because there is a copy button and might be a response to someone's ticket and you might save time here and there. So alerts we covered, device-specific page, you will also see the copilot button and you can start by asking something that, I don't know, maybe something about my laptop. Like what's the serial number? Let's see. So something that can be very device-specific. Copilot will go and look for this information. Maybe you can ask to list certain processes, like do something that is, yeah. So this is exactly what it is. We can go to the device itself and you will see the serial number, which is absolutely accurate. So those are the simple use cases that how you can put the copilot, like into good use, however, the magic starts when you start tackling your day-to-day tasks. So this is all cool actions, however, how to make it more meaningful. Because at the end of the day, we started our today's conversation with my story about no one actually likes to type. So, and as you can see, like if we open any given ticket, so we have like our main UI. However, at the very bottom, we have additional field, which is called ticket properties. And the ticket properties, depending on the use case, might be very, very different. Let me put things into perspective. So in our admin section, so for whoever is new to the product, in support and ticketing, we have the ticket forms. The ticket forms is basically a template, which has a predefined subset of custom fields. And until almost like the most recent times, it was pretty much manual. So you can put different types of custom fields. It can be free text, it can be a dropdown, can be dropped down with dependencies, it may be a calendar picker, can be many things. But it all ends up with custom fields. But what if AI would fill them in for you? And the spoiler, it is possible. That it's not only filling in those fields, but transforming the data you get. And it will get that to the point that it will fill in, like, for instance, like your actions. Maybe you want not the static value of like, okay, I did this troubleshooting steps, but then I needed to get back to the ticket and do something else. How to make it dynamic? How to stop forgetting updating so many fields that are required for your internal reporting? How can you justify the complexity to a manager? Does this ticket require a license to be purchased? Like yes or no? Is it about cybersecurity? In the resolution summary, why should I type it if AI can do this for me? So this is where it all starts to come together. And as you can see, I've selected the basic form. And let me show you where it lives in the product so I can continue to the more complex part. So whenever I select the basic form, I will get a subset of custom fields. Some of them are mandatory, some of them are not. But as soon as I select them, as you can see, most of them are empty. Okay? However, I need something to actually trigger AI to be intentional about filling them in. So for this, I can go to ticket automation rules. And you can go to like the ticket. So let's see. So whenever the ticket status has changed, like for instance, you're changing like the ticket from the open to pending, you need the AI to check the answer to the question. So is it about cybersecurity? Yes or no? Does it require a license to be purchased? But in this specific use case, it's one. But in this specific year... However, we can new ticket rule. So let's go and create something for the new tickets. So the new ticket went into your like ticket queue. Let's create like, so we have a condition that it will follow. We don't require anything else in the condition section because we have the trigger. So for all the new tickets that come into play, we'll do something. First of all, we need to trigger somehow the form which will bring in all those fields which are relevant for this type of the ticket. Maybe like, for instance, like if the ticket title contains Wi-Fi, then trigger like this or that template which is like for wireless tickets only. In our case, we'll just use the ticket form and for ease of demonstration, I'll just use the basic one. I'm setting the ticket form basic and after it's a basic form for ease of like AI actions. Well, I don't know, intro. I don't know how to call it. I'll just call it like this. Okay, so whenever in the new ticket is created, I will go in and so if the ticket property, by the way, ticket form equals basic form, then I will set the field value by copilot. So actions been taken by the technician so far. So set the field value by copilot. Maybe like resolution summary, bulleted list, maybe something on top of it. So maybe I wanna know, is this ticket a cybersecurity? Is it about cybersecurity? So whenever a new ticket comes in, by the way, they need to be in this order and then when I want the troubleshooting to be more meaningful, ticket status change, let's say ticket updates. We'll ignore the flow and every time the ticket goes from open to pending, from appending to result, some fields should be filled in. So the set field value by copilot, I will do just the actions taken by the technician so far. Bear with me, I will show you the real ticket example. I just want you to see the setup, how it works so we get to the actual point and whenever the ticket is resolved, let the manager know. And whenever it's resolved, I wanna set field value by copilot. I want the resolution summary once again to be filled in and I want the complexity to be justified to my manager because I don't feel like I'm filling those in. Okay, I have these four rules and now let me try. So as you can see, whenever a new ticket comes in, it should be set to the basic form, which is the foundation of the ticket automation. To demonstrate this, I have a real laptop and let me actually start, not a new chat, but a new ticket. So I'm touching the autopilot realm because if we would have an autopilot, you can start with a go-to AI system then it will start dealing with the end users issue so you don't have to. So I'll select the basic form, like if I don't have 20 something in my demo environment, something's wrong with my antivirus. I'm trying to make it as close to real life as possible. Maybe you had those tickets, please help. It shows some red alerts. I don't know what this is. Sorry for my typing. And also it's a remote session, so. Okay, I started a conversation and now I need to see whenever the ticket comes in. Yeah, there's the ticket. It's not yet assigned to anyone. However, the basic form has been applied and everything seems to be still empty. The SLA was applied, which has nothing to do with what we're trying to achieve. Let's wait a couple of seconds, but I don't think something else should happen. Okay, and let me save time by generating a reply. So instead of me typing this amount of words, I can just hit send. And like as an option, I can just hit pending. So I could have pushed it when I was responding to the ticket, but I didn't. Okay, the field seemed to be still empty. What does it have to do with the IT data? Well, now you enter the realm of the data, like, okay. So a technician requested a screenshot of the red alert message. An antivirus issue reported. User sees red alert. Assistance requested. Now I will try to emulate that I'm an end user. I think it's something about my expired subscription, but I don't want to pay for the antivirus. I'll make it as funny as possible. Not the most exciting ticket to work on. However, in the background, you might've noticed that in the ticket activity, the ticket went from open to pending. The system started to do automation. So the gear sign symbolizes that the ticket automation kicked in. Now the end user has sent me the ticket. Now I need to do something to respond to it, okay? So I can, once again, click on generate the reply. If you don't want to renew your screen, you can uninstall the current antivirus. So might not be the relevant response. So, but I will like, I fixed everything. I'm a superhero and I'm typing this manually, but I actually, I'm a superhero. Your issue has been resolved as I went to the admin panel and did everything for you. You're welcome. Okay, we can, doesn't sound like a professional IT response, so we can do more formal. And as you can see, that my completely inappropriate language turned into something that is more professional. Rather than having quick reply templates, and I can just set it as resolved. And now, as soon as it's resolved, you can see that some of the fields are still empty. For instance, like, let me, yeah, let's, nope, so it actually doesn't. And I still have some empty fields, like justify the complexity to the management, but it only applies to resolved tickets. It will stay empty until it's time. Let's see. Okay, so the actions taken by the technician that requested a screenshot, and issue resolved by accessing admin panel and adjusting settings. Is this ticket about a security, like was not picked up, but we'll check this. So the resolution summary, user reported an issue with antivirus and red alert, technician requested, user mentioned, technician resolved, I'm editing. I'm editing an internal note, time saved, and the day is also saved, probably. So how do we scale this? This is one of the most important questions, and this is where the interesting happens. So we have section in there, which is reporting. We have two types of reports. One is like the operational, one is analytical. Operational reports is not as customizable. It's hard coded. Most of it is just filtering options, no customizations whatsoever. On our master enterprise for ITDs and power and superpowers, subscription plans, which is tier three and tier four, respectively for both products, we have the ability to customize the analytical reports. What does it mean? You can have access to data set, and you can start to get immediately so much more information about like something that is very specific. Let me even, I don't know, let me do the devices data set, but I still wanna use the, I'm trying to make the use case a bit more complex. So I will start to build a report, which is like ticket ID. Oh, sorry. Ticket, let me see. Let's do the ticket title. Oops, sorry, wrong data set. Let me do the tickets instead, because not all of the ticket related information present here. Ticket ID. So I'll go with ticket ID. Ticket title. I can add some dates. I don't know, just when the ticket was opened. Ticket. Ticket created date. Status. So ticket activity status. And now let's add something that is account related. So as you can see, we've added a bunch of custom fields that I want to tell my manager that I want to, like, hey, this is why it took me so long to respond to this ticket. So, and maybe you're dealing with, like, you're in MSP and you're going to your MSP customer and trying to justify why their block hour contract was almost depleted this month. Maybe, like, the resolution summary needs to be added because it's something that you need to be compliant with and you, as a manager, you need to check your IT technicians that they fill it in. All of it can be done. As soon as you hit run, you will see, like most of the tickets, they will start with the smallest one. Give me just a second. Okay. I need the most recent one to show you that the data is loaded. Yeah, so as you can see, all the tickets that have changed status since I started to build this report, those fields have been filled in. And this is how you can get the ticket to scale. The tickets that were created before this, they were not affected. But you can change something in them to trigger the change. And this is how, like, for instance, you need to explain, like, 160 tickets because you bill by the ticket ID. And this is where it can be, like as an MSP, it can be extremely helpful. Maybe you want to find a specific keyword that like the resolution summary contains, like let's do the summary, contains a specific word. You want to know something about Zoom. You just type it in, hit run, and you'll see the tickets that are just about Zoom specifically. And this is how we can find something that has never, like, been out there. But now you do have it. You save time, you type less, and it's all good stuff. And it's just one of many, like, simple use cases. We don't want to overcomplicate it this time. However, how can you make AI to do a better job? So on the ticket, whenever it's resolved, let me do the just the resolved tickets. So the ticket was resolved and I have the generate article button. I click on the generate article. And next time I'm dealing with the same type of issue, I will have an almost like ready to go KB article, which will be generated in a couple of seconds. And I have the troubleshooting steps. And you can approximately understand that it's like, I don't know, half an A4 page worth of text. So it will take you a decent amount of time to type it in. And maybe you will do it undistracted, but more than likely you won't likely you won't, because someone will come to your desk, because this is the nature of our because someone will come to your desk because this is the nature of our. profession, being IT professionals. And as soon as you do it, you can make it as internal and external. For MSP specifically, very often overlooked, you can do it as site-specific or customer-specific, so you don't have to share the overlapping knowledge, or maybe there is something that is like very customer-specific, which should not be seen by anyone else. So this is where you I got many questions on this one. Okay, that probably concludes the demonstration. We have some time left for questions to be answered, and I will be more than happy to. Let me stop sharing. Thank you, Dima. I think, again, today the purpose was to be very, very specific on those niche, yet very powerful use cases, where we can leverage AI to automatically populate content in your case, you talked about custom fields, by actually reading some of the data within the ticket. And you went ahead and you showed a use case leveraging the end-user portal. So you went into a lot of detail, but I want to make sure that we're really simplifying this specific use case. There was a question from someone in the audience, that what if AI made a mistake? What if the technician made a mistake as they're responding to a ticket resolution? Does the AI learn from that, and is that what's applied later? So AI is learning from the existing KBs. This is how you prevent this from happening. Also, you can create a custom field. I never did it myself, but we probably can try. Track technician's mistakes, create a custom field to do exactly this. So if there are any mistakes to be tracked, maybe you can actually point them out for some coaching purposes. Okay, and then Jeff here is asking us, you know, can we somehow message the user? So you showed the use case of autopilot, end-user portal, and the Atera ticketing side. Can we use message users in real-time using MS Teams, or WhatsApp, or some other way outside of the customer portal? Okay, so the short answer is yes. A bit longer answer depending on which features you have, because for autopilot specifically, we have the Slack and Teams integration. We also have the coming up WhatsApp integration. No set in stone date yet, but we will introduce it pretty soon. By default, the entry points to initiate a ticket is the portal, the email, the tray icon. This is where your end-users can start the conversation with the technician. And on the portal, they will see all their open tickets, which is convenient, but the messengers is something that lives in the autopilot realm, but it will give your technicians more entry points to communicate with the end-users, which is cool. But at the moment, it is autopilot specific. Okay, and I think Jeff is really into autopilot specific today. Thank you for your questions, Jeff. Even though we didn't focus on that, but I still want to respond to it. Can we set up autopilot to send a KB article to a client when a ticket is opened? Yes, we can. And actual autopilot will do exactly what copilot does. So if you ask it a question and the knowledge exists, it will pinpoint the KB article which end-users should be referring to. In the autopilot case, on top of the knowledge-based article, there are custom instructions, company description, company terminology, and company software stack. So basically, if you're not allowed to use certain pieces of software, well, too bad. If you want the Adobe Creative Cloud license, but you work in finance, probably not a good idea. And we're not done yet in this realm, so stay with us. And we have more cool webinars on the autopilot topic and copilot too. And it's great that you bring that up. And Jeff, I do encourage you. We run IT autopilot specific webinars every other week, and we focus specifically on those use cases. So I invite you to join those. And we can actually demonstrate some of these use cases. To our audience, I'm more than happy for you to ask questions. I know, again, we focused on a niche use case, both leveraging custom fields, using those custom fields to populate, and then in the reports, Dima, as you showed, leveraging AI to populate information that you can later use within those specific reports. Dima, many times, and again, we didn't touch on some of the broader and most common use cases of AI Copilot, whether it's creating scripts and leveraging that. Maybe you can touch a little bit on what you're seeing, let's say, additionally within AI Copilot that clients are leveraging. So the scripts are a great example. So creating scripts with AI, because no one knows PowerShell end-to-end, no one knows Mac terminal end-to-end, and you can create OS-specific scripts. You can also do sandbox testing for those before applying them to anything by your test device, which is very cool. And also, not only creating them and, like, leaving them in your script library, you can allow them to be used by the AI. Copilot will, like, for instance, Copilot will find certain files in a certain folder, but by our security policy, it's not allowed by design to delete those files without your permission. And to give this permission, you need to inject a script. So you can go ahead and delete the files in a specific folder if being asked to. This is one of the examples of, like, using the AI responsibly, because we don't want this to go absolutely crazy. This is one use case. And so the scripting can also be, like, end-user-specific, it can be customer-specific, it can be technician-level specific, because the scripts are not only, like, created, used by AI, but also they can be segmented by the technicians, their custom access rights, and it actually, the list goes on and on. Also, the AI capabilities live in the PowerShell realm, so the command line tools, they all have it, and you can save some time typing. Again, we're getting back to the most basic thing that we do with a computer, we type, and this is where we can save a lot of time, resulting in a lot of money savings, and actually doing meaningful things instead of doing things that you perfectly know how to do, but sometimes you have this friction of, like, just starting the actual thing. And this is where AI saves our lives. It's not here to do something bad to our day-to-day work, it's there to augment our performance and make it better. So why not use it? Me personally, I use it all the time just because I love voice input because it's just faster. Thank you, Dima. And we have some questions here about the business model that I'm happy to answer. So if you're new to Atera and you're just trialing it out, then you have AI Copilot available in your Atera trial for 30 days. All of the capabilities open for you to test. If you are a current Atera customer, you can subscribe to a 14-day free trial and test out all of the capabilities, some which Dima demonstrated today. In terms of the IT Autopilot, that is a separate product. It works very closely with the Atera core platform, but it is a separate product. And if that is of interest, as I said, you're welcome to join our webinars or to reach out to your dedicated account manager who can help you set up a trial to test out IT Autopilot, which is the end-user personal AI assistant that you can leverage. But Copilot is available to your technicians within the platform. You can easily start the 14-day trial. It is a chargeable add-on per technician. I hope that answers your question, Mark. Okay. And Jeff, I hope we did answer that. And there is a question here whether the webinar is recorded. Yes, this webinar is recorded and we will be sharing it. I think within 24 hours, you should get access to the recording and the presentation, as well as the AI Copilot playbook. Last but not least, Mark mentioned in the chat, you mentioned two types of AI tools. Is there a charge for both? I'm currently subscribed to AI Copilot. So the two AI products are separated for a reason because the audience by design is different. The Copilot is designed to help your technicians. So the Copilot is paid per technician in your account. So it's immediately available to all the technicians. So if you have five, you pay for five. ITO to Pilot priced against the end-users that you're dealing with because the main person who benefits from it is the end-user. So it has nothing to do with a technician. I don't know, maybe besides the setting it up, but the end-users are the ones who benefit from having the 24-7 support, tier one assistance and whatnot. Thank you for that. And Dima, any final words that you'd like to mention? Words of wisdom. Words of wisdom. The only thing that I would like to encourage people is to be not afraid to use the tools because it's the only way to actually click around to find out in a good way. Because if we don't use it and if we do bad prompts in our day-to-day, like I know maybe like two primitive prompts, we will never get good results from AI. If we like try AI tools here and there, we try to augment our daily life, our daily tasks, maybe like we want to tweak our nutrition, maybe we want to tweak something else, which is completely not related to work. But if we will understand how it works, we'll get more educated on how to write clever prompts, we'll get better results. Because the better the prompt, the better the output. So I encourage you to experiment with AI tools you're using. Try AI Copilot. Reach out to us if you want to try Autopilot too and see how they work all together. And that will be probably my final words for today. Thank you so much, everyone. Thank you, Dima. And again, I want to thank our audience for attending today. We appreciate you joining and we encourage you to sign up for our additional series where we can help share additional knowledge. And of course, some of the key use cases that customers are doing to leverage AI. Thanks again. Thank you, Dima, for that detailed demo and hope to see you all in our next webinar. Thank you so much, guys. Yeah, have a great one. Bye-bye.
