Ep.147- AI Agents – What's new and interesting in AI, and what does this mean for you? With Sam Nelson
Subscribe to the Next Level BizTech podcast, so you don’t miss an episode!
Amazon Music | Apple Podcasts | Listen on Spotify | Watch on YouTube
Transcript is auto-generated.
Welcome to the podcast that’s designed to fuel your success in selling technology solutions. I’m your host, Josh Lupresto, SVP of Sales Engineering at Telarus, and this is Next Level BizTech.
Hey, everybody, welcome back. Today, we are talking about some exciting stuff. We’re talking about AI agents, what’s new, what’s interesting in AI, and what does that mean for you? And we have on with us the amazing Sam Nelson, VP of CX at Telarus. Sam, welcome back.
Thanks for having me, Josh. Always excited to be on the podcast.
Sam, we got–
man, I feel like not a minute goes by where we’re not sharing some new, latest, greatest thing and trying to decide what this is. And I figured, we’re talking about some of those. The partners need to hear about it. So let’s jump in. Before we do, I want to remind everybody that Sam was on episode 108, the beginning of the year. We were kind of just jumping into some of the AI trends that you shared,
things that was really kind of starting to elevate the CX experience already. So if anybody hasn’t seen that, go back and listen to episode 108. And then also, you’ve got a great production out there every week that comes out, the minute snippets. I think it’s crazy the amount of things that you’re able to pack into that. So if anybody has missed that, go check those and keep watching those every week. I think you put some great stuff out there for our advisors.
Thanks. I can’t believe it’s already been so long since we last spoke. It’s crazy.
I know. I know. And there’s still so much to talk about.
So much has happened.
We’re going to jam as much of it in as we possibly can. So we talked a little about your backstory, so we won’t go into that. We’re going to make everybody go back and listen to episode 108 or one of the others that were on there. But just give us a start us off with hard lesson learned, mentor, mistake, old boss, anything entertaining. Just start us off somewhere.
Anything entertaining. I love it. You know, what’s funny is I am super late. I was super late to the Ted Lasso bandwagon. And so I recently watched that. And the one quote that stood out to me because there are so many out there. Great series. If you haven’t watched it, 10 out of 10 recommend. But the one quote I took away from that was don’t fight back, fight forward. And that’s one of my favorites when I think about it. You know, every all these distractions and everything going on and we tend to react and, you know, ways that don’t progress us forward sometimes just naturally as humans. Right. But if we focus on eye on the prize and we move it forward, we can get a lot farther with a lot less stress. So definitely something I’m taking into the new year.
I love that because I suppose if you don’t look back, you don’t have to look and see what everybody else is doing. It doesn’t matter. Just focus on you and focus on the forward. I love that.
Exactly.
All right. Well, let’s let’s dive in. We got a lot of got a lot of topic here. So we’re talking hot trends, AI, which is we got all kinds of stuff. We got AI agents. We got this anthropic computer use thing that’s floating around. Let’s let’s level set here. And first of all, you know, we know the word agents from traditional contact center side of things. Right. We know what an agent is. But first of all, just to find for us, though, what is an AI agent?
Yeah, for sure. That’s a great, great point you make there, Josh. I think a lot of folks just to level set with everybody, whether you’re selling CX or not is a lot of people consider an agent to actually be a human sitting in a cubicle answering phones, right? Answering calls now answering emails and chats and things of that nature. But in the context of artificial intelligence, an actual AI agent is some kind of either system or a program that’s actually designed to autonomously perform tasks on behalf of a user or doesn’t even have to be a user. It could be a completely different system. Right. And so essentially it’s this mechanism that can make decisions. It can plan. It can interact with different environments in order to achieve a particular goal that you give it.
So just know that it’s not necessarily human. Again, it’s a system or a program designed to do things.
Awesome.
So we’ll get into anthropic, some topics for today. We’re going to talk about what anthropic is doing. We’re going to talk about what Salesforce is doing. But maybe, you know, just since now, from the last time that we chatted, we’ve got this evolving landscape just moving so quick. What are some of the most significant changes that you’ve seen in the market with some of these introductions of the AI products in the last nine months?
There are so many. It’s too wild. It’s like, where do I even start since episode one away, Josh?
There are so many different use cases. And what’s interesting now is we are starting to see more AI use cases specific to industry or vertical.
And all these new companies coming out with different features, leveraging different kinds of LLMs or large language models, we’re starting to see more of that come up past the basic, hey, analyze this message. Right? We kind of started with analysis and insights into data. But now it’s how do we act upon that data and the AI is actually helping us do that in this next step of AI evolution.
So let’s fast forward then anthropic, right? You know, when we talk about the big LLMs that are out there, we’ve got chat GPT, we got Microsoft Co-pilot, we got anthropic. And so there’s a feature that’s come out, right? If we understand, I can chat with all of those LLMs, I can ask them for advice, I can, you know, it will generatively reply and give me suggestions to anything from what do I do here to help me build a strategy with this to, you know, help me do anything that I want, write code, you name it, everything in between. But now, anthropic has come out with this thing called, which I would have loved a better name on this, but it is what it is. It’s called computer use. So it says, okay, I, you can do an API into a virtual instance, it sounds really confusing. So I went and played with this. And I think, oh my gosh, this is going to take forever. I’m, you know, I’m not a JavaScript guy, I may be a nerd, but I don’t, I’m not, you know, that deep in writing code. And it was really simple. I take an API key that anthropic gives me, I plug it in over here into a virtual machine. And poof, I have a little chat window that now I’m not just chatting with the text base, I’m chatting. And I’m watching this thing, hey, go book me a flight from Salt Lake to LA on Tuesday. And I’m watching it open up Firefox, I’m watching it pull up times, I’m watching it look at schedules. So we’re moving towards this idea of full blown autonomous, right? So that’s, to me, this is a not highly touted, but monumental leap from an AI interaction perspective. So as that’s moving along, what does this mean? How does this play? How do you want advisors to think about this?
Yes, there are a lot of crazy cool things coming out with AI in the realm of AI, especially with computer use. I’ve seen it, I’ve tried it. It’s really, really interesting. I think it has a long way to go in the, in the context of what you mentioned, look up flights for me, right? It can look up flights for you, but it won’t take it a step further and actually book things based on your preferences. You still have to provide some kind of decision making there, right? And I think that eventually we’ll get there, but it needs to still have that human element of knowing what it is that you want to do, right? And so I think that when we look at practical use of it right now, it’s mainly around the research, right? And cutting down the time and increasing overall productivity and how AI is helping us do research. But this poses an interesting sort of conundrum with our technology advisors, right? Is how do you have the conversation around these things that are new really to everybody, right? It’s even new to us. Josh, you and I were just talking about this on the other day. Check this out. It’s brand new. Let’s look at it. And the reality is clients, businesses out there are going through the same exact thing. And so as easy as it may seem, right? Plugging in API keys and things and such. I’m not technical either.
But there is absolutely a technical component and other components you have to consider to complete the tech stack. And for tech advisors, it’s providing that word of caution to business users of, “Hey, that’s really cool.” But we talk about data security and data cleanliness and things of that nature, right? If you are leveraging AI, it’s only as good as the data that you provide it with and that it’s being provided and the sources it’s being provided.
I think the value, just thinking of this too, is that if our partners are going to get into these conversations and they’re going to dive deep and they’re going to uncover, right? And we’ll talk about some of the things that you’ve done at some of the latest AI summits and examples.
But again, I don’t want to miss out on the underscoring of, okay, we explore this stuff because we’re passionate about it and we love technology. But what you should also understand is that we can come in and help you. We’re so far in front of some of the real commercialized productization that we’re here to help you dive in from a technical perspective, engineers, architects, advanced solutions, whoever you need to bring into that call, whatever that layer of the discussion is. Don’t forget that, yes, we’re passionate about this, but there’s a purpose at the end, right? We’re here so that we’re ready for whatever it is you all need to pull us into.
Yeah, absolutely. Don’t go in alone because I guarantee that that small AI comment or conversation you have is going to turn into something much bigger.
Yeah, it seems like this is one of the greatest opportunities for technology sprawl and runaway IT, just like we thought cloud was this scary runaway IT thing. And now look at how we’ve been able to monetize it, partners have been able to capture that. I think this is that next thing where, hey, this can be great if you do it right and to do it right, let us help you with that.
Right, right. Yep.
All right. Salesforce,
we’ve sat adjacent, we’ve integrated a zillion CX environments into that. We’ve maybe consulted of how to build, what to build, things like that. But Salesforce has put their foot down in the CX landscape with this little thing called Agentforce. Talk to us about what that is, what’s its role in CX and what’s your thoughts?
Yes, it is a big move for Salesforce. And as we all know, they’re a behemoth when it comes to CRM. And so this is really important for technology advisors to understand is a lot of the CRM players, not just Salesforce, but even Microsoft in the world of dynamics,
and some of those other larger players out there are actually looking to enhance the agent desktop. And so essentially what that means is they are now giving consumers, yes, everyday consumers, the ability to create these AI agents within their solutions, within their CRMs. And so you know, having that capability sit in the CRM is extremely powerful because if you’ve ever used one, and we do every day, you are living and breathing inside that thing all day long, right? And so the capability is there. It’s there, but it’s the fact that it is now available to consumers. It’s actually been there for a really long time. If you think about it, right? Automations,
you clicking a button and it does something, right? But now you have control as a consumer to define exactly what happens by just providing it a bunch of data, giving it some parameters, and off you go, right?
But again, what’s really important to understand is it does still require a lot of technical work. I mean, Josh, you and I have tried this. Yeah, we’ve actually tried this in Salesforce, just in a random environment. And it’s not easy, not yet, right? And you have to consider these out-of-the-box solutions that we have in the portfolio today at Telarus make all that stuff really easy. And so we’re at this kind of pivotal moment of, you know, we’ve got these big behemoths coming in saying they want to take over agent desktop, but at the same time, we have to help them understand that that’s not always the easiest route unless they have resources. And we need to be a part of that resource conversation. In fact, technology advisors need to position themselves as resources for these new features and these new AI agent capabilities coming out from these behemoths.
Yeah, you know, it was a fun, interesting exercise to go through. And I think we bring it up so that our advisors obviously know what’s out there, what their customers, what their prospects are looking at. But the reality is, if you are the CIO, if you’re the CEO,
do you right now want one person that may be the expert building that and being dependent on your entire infrastructure being built on that? Or would you rather have a full-fledged, mature platform that is ready with a huge wrapper of team around it? I think those are the things, you know, we’ll continue to pay attention to this as this evolves. But it’s compelling. They’re making an effort and they’re going to be in these conversations. And I want our advisors to be prepared that they’re going to be in these conversations.
Right. And on the bright side, honestly, Josh, it’s good to bring this stuff to the forefront, right? Because it actually puts CX
at the top of everyone’s mind. And so if anything, now is the perfect time to start talking to businesses about what exactly happens on the agent desktop and when, because you can help them go down that path and help them define what that experience looks like without them just kind of experimenting with something on their own.
All right. So Thanksgiving is over. We’ve all had our turkey. We’re looking forward now to 2025.
We’re thinking about AI CX 2025. So your role as the VP of CX, what’s your read on how these businesses are going to approach that in 25? And then what kind of trends do you see being implemented or any examples around that?
The crystal ball. Who are favorite? That’s my favorite question.
It’s Cleo.
Yes. Yes.
Well, first AI agents is going to be a huge theme. Really, I see just a couple of themes going into the new year, right? Because we are, again, like I said, at this pivotal point of where we, as well as customers are still trying to understand the capabilities of AI. And it’s actually moving a lot faster than we thought and that we can understand. And so going into the new year, AI agents in the context, again, of artificial intelligence being a program or a software or what have you, that’s going to be a big theme going into the new year. People are going to wonder how to best either create or use AI agents to optimize business processes, things of that nature. The second trend that I’m predicting here is around industry or sector specific innovation. Like cool. Sounds great. What does that mean?
It really means that we’re going to start seeing more and more capabilities emerge based on industry specific use cases.
So think of it as, you know, say in healthcare, patient experience, booking appointments. What specific solutions can you leverage to book appointments versus a solution boiling the ocean saying let’s book appointments for everybody. It’s going to just get really industry specific and continue to grow in that fashion as we go into 2025 because every single industry’s use cases are just being hyper focused on right now.
Love it.
Is there anything additional there that you think, you know, is personalization, is that same as vertical specific?
Any differences in that or are those one of the same?
Yeah, that’s a great question. I think it is a little bit of both. It’s one of the same and different because I definitely think that when you look at patient experience versus a retail experience, right, they’re going to look dramatically different. However,
hyper personalization, leveraging artificial intelligence is not going to slow down. So I would say that’s almost a sub trend that we’re going to see sprinkled into everything. People are not just going to look at, hey, how do you book appointments with a primary care physician, but they’re going to look at, okay, you know, based on a condition or your time preference or date preference or location preference, how do they further tailor booking appointments to be a better experience for you as a patient. So it’s definitely sprinkled in. It’s not slowing down and it’s going to be interesting to see what more we can personalize out there with AI. The fact that we can even just generate videos without even, you know, engaging other humans, it’s pretty wild. I’ve seen too many cat videos.
Yeah, I know we say this a lot right now, but there is no funner time to be this Jim Gaffigan reference. If anybody can remember what that’s from.
But there is, it’s just, this is so exciting to see and just think about how can we apply this? How can we use this? How can we help our advisors? There’s just, there’s no shortage of lack of innovation and then figuring out a use case to apply it. And what we have to caution, I think to put a bow on some of that, what we have to caution the advisors on is that the customers are, you know, the IT people at these orgs are equally as excited, right? And so as we look at some of these, are we looking at, and you know, let’s say the customer brings up, “Hey, do you have these 95 things I’m looking at?” Are we a problem looking for a solution or are we a solution looking for a problem? Because the first is a lot easier and ROI’s out a lot more solid. And those are interesting conversations to, you know, call somebody’s baby ugly. But I think we have to frame it up and try to decide which product fits where.
Yep, absolutely. You have to.
Okay, so this is all rosy and exciting.
But, and I think we make it sound really easy sometimes.
But now we’re introducing really, really powerful technology that, you know, I think it’s, it’s, there’s APIs, there’s integrations, there’s data cleanliness. I mean, talk to us about, while it’s got all this promise, what are some of these difficulties that you’re seeing the companies struggle with in trying to do some of this on their own? We touched on it a little bit earlier.
Yeah, I think the number one thing is, like you mentioned earlier, they see something shiny and new. And they think, okay, what can we do as a business to accommodate for something like that, when in reality, it should be the way around. And I think we don’t consider, as well as customers, don’t consider the security concerns around using things like chat GPT, like clog, right, we don’t consider the information that we put in there, we think, oh, this is so easy, let’s just pop in everything. And next thing you know, you’re leaking a bunch of proprietary data into the ether, right? And if it’s not blocked off, that’s a huge problem. And so I think that, first and foremost, there is a security concern around implementation and actually using large language models out there. Everyone wants to use the most powerful, but it does come with security risks, if you don’t handle that appropriately.
And on top of that, around implementation, you hit the nail on the head with customers can only leverage artificial intelligence, as long as the data is good. And so that becomes a cloud conversation, data integrity.
That’s I would say challenge number two is making sure that data is clean, because if it’s not, you’re going to, you’re going to suffer when it when it comes to AI and its outputs.
So, so if I’m a partner, if I’m an advisor, and I know that my customers, this is the questions section, I love questions. So if I’m an advisor, I’ve listened to what you just said, I know some companies are going to be struggling with that. And I think maybe my customer might be going through that, but I’m not quite sure what to ask them. What’s your, what’s the one question that eases them into asking that would elude out then a, oh, I hadn’t thought about that, or, oh, you know, security, wait, what, what’s your, what’s a good question to ask there to uncover, are they going down that road? Are they have they thought about this?
My favorite question, and I will totally steal this from our amazing Jason Lowe, CX solution architect, it truly is what is your AI roadmap going into 2025? If you ask that question, it will expose quite a bit, they will either say, and most likely they won’t, but they will either say, here it is, which is very rare,
or the latter, which is, what do you mean an AI roadmap? And that is the perfect place to start. So again, just to reiterate, what does your AI roadmap look like for 2025? And that will set off so many light bulbs. I mean, just think about getting in your car and all the emergency lights are on, right? It’s going to inspire that feeling for an end customer to consider, because I guarantee that they’ve touched artificial intelligence in some way, shape or form, right? We can, we can for sure say that. But now you want to unpack exactly how much and what they’ve considered around how to use AI in the business. So that’s the number one question. And usually after that, it’s okay, well, what do I, what do I do now, Sam, I’ve asked them, right? Talk about AI readiness. And that’s where you have the conversation around the data. And we actually have an AI readiness document out there. So yeah, we’re going to actually continue to build more resources around that. Because what I should have added this to my third prediction, but it’s this born in AI, quote unquote, born in AI consulting, that has now emerged. Because if you start with AI, and you start with a conversation of how do you want to leverage it, let’s look at your use cases. That leads to all of the different technology swim lanes, not just CX, but cybersecurity, cloud, advanced networking, mobility, IoT, right. And so it’s a new, different way of actually selling, but it crosses over all the different swim lanes. So very powerful.
I love it. You mentioned cars,
car mechanic back in the day, my this there’s a total parallel here, people would come in for like, I need my oil change, right. And you know, obviously, our goal was to, what else can we find that we can help them with? Besides this one thing, we’ll do this one thing, but we’ll say, did you know that all these other things are going on on your car? And my favorite, favorite, I saw this more than once more than twice, probably more than three or four times. You mentioned all the lights on on the dash, you know that if you take a little piece of black electrical tape, and if you’re at the driver’s seat, and you put that tape right over the check-ins, and like, you don’t have to see it. But it doesn’t mean that it’s not still on. So don’t let your customer be the the tape over the check engine light person. And I love your idea to how we can help get in front of it. Love that question. Awesome.
Okay, final couple thoughts here. So I want to move into a little bit of practical advice, and then we’ll get into kind of some some wrap up and takeaways here. So final thought on practical advice. So if partners, okay, we’re getting them warmed up to the idea of this AI and CX tech sale.
What’s the practical advice? You’ve done some great stuff at some recent events, but what would you help them to suggest to initiate some of those meeting conversations, meaningful conversations that kind of capitalize?
Yeah, when it comes to customer conversations, I say, you know, start with the general questions, don’t get too specific, and really try to understand a use case. When I have a client who comes to me and says, I want to use AI. My next question is, well, how do you envision AI helping you in the business?
And there’s all kinds of different things that I get. And you could tell that they’re just going in all these different directions. Help guide the conversation. But more importantly, bring a resource. I mean, that’s why we have three dedicated CX solution architects and a huge, huge field solution engineering team, right? Don’t go in alone, bring us along for the ride there. But Josh, I actually wanted to talk sort of more practical use, right? Because I think a lot of people are thinking, well, how do I use this in the business? And you talked about AI, our AI summits that we had pretty recently, we had three of them this year. And I actually went on stage and demoed a couple things. I demoed GPT, chat GPT, specifically, and I gave it some prompts. And you have to understand that prompting is not just, you know, hey, tell me about this. It’s a lot more complex than that. You have to give it a lot of parameters.
And so we covered that. But what’s even cooler is I think advanced voice mode. If you haven’t tried it yet, try it. Advanced voice mode with chat GPT.
One of the practical use cases is if you’re one of those who’s a talker versus a typer, this is going to help you significantly. You could be driving in the car and having a full mind mapping or brainstorming session with advanced voice mode, and it will generate a dock for you afterwards of a summary, right? People are doing things like writing manuals on the go.
I tried this the other day. I walked to my dogs. And I had a full conversation with advanced voice mode around some ideas going into next year. And it was brainstorming with me. And when I got back home after a 15, 20 minute walk, it had a document ready for me summarizing everything we talked about and some action items that I should take from that. So definitely, you know, leverage these tools yourselves because the more experience that you have with them, the better that you can speak to them in front of your clients. And honestly, even if you can’t monetize something like that today, the fact that you just bring general knowledge around productivity tools, that speaks volumes.
Yeah, awesome thought leadership. I love it.
Great, great advice.
Okay.
All right. So let’s say key takeaways here before we wrap this up.
What do you want? We talked about a lot of good stuff today, but what do you want the audience to remember from today about AI agents and how to roll that into, you know, the next nine to 12 months?
Yep.
I would say number one thing, when you ask a question, again, reiterating the question, if you’re tuning in towards the end of this podcast, it’s what does your AI roadmap look like for 2025? That is the number one to open the can of worms. And then two, just keep in mind the trends, AI agents plus industry specific. So when you look at your client base, or you’re prospecting into, you know, new client accounts, consider the different use cases before you go in so that you’re fully equipped to have the conversations to take them to the next step.
Love it. All right. Final thoughts here. Take this anywhere you want. Anything you’re excited about yourself over the next 12 months, are you going to buy a robot? Anything else you want our advisors to pay attention to? Off topic, on topic, take us home.
Look, man, I have a Roomba and I’ve had a Roomba for some time. And that thing is just, has changed my life, but it’s already been around for some time. If you think about it, right? Now, am I going to go buy an Optimus? I’m not sure. I’m a little worried about kind of the iRobot situation. But assuming we have the parameters in place, I mean, hey, you never know, we’ve already got self-driving cars. So go figure it’s the next step. And we just got to keep up.
Love it. Love it. So much stuff to pay attention to. Great stuff. Appreciate you coming on. As always, Sam, thanks for thanks for dropping all the knowledge and doing all the things that you do.
Thank you. Appreciate it.
All right, everybody. That wraps us up for today. I’m your host, Josh Lupresto, SVP of Sales Engineering at Telarus. This has been AI agents, what’s new, what’s interesting in AI and what that means for you, Sam Nelson, VP of CX at Telarus. Until next time.
Next Level BizTech has been a production of Telarus Studio 19. Please visit Telarus.com for more information.