Ep. 109 AI Overload: The Mind-Blowing Ways Artificial Intelligence Elevates CX! With Dan O'Connell of Dialpad

Subscribe to the Next Level BizTech podcast, so you don’t miss an episode!
Amazon Music | Apple Podcasts | Listen on Spotify | Watch on YouTube

Listen in today as we talk to industry-leading expert, Chief AI & Strategy Officer of Dialpad, Dan O’Connell. We first learn as Dan started at Google when it was sub-300 employees and how that journey has progressed into an acquisition that propelled him into this role at Dialpad, where he’s managing not only Strategy and AI but must distill down what the products are that market needs. There’s so much on AI and CX that you don’t want to miss all the ideas Dan drops for partners!

Everybody, welcome back here. Today, we’re talking about the topic that everybody wants to hear about. We’re talking about AI. This is titled AI Overload, the mind-blowing ways AI will elevate CX. Today, excited to have on with us, Mr. Dan O’Connell, that I think has one of the coolest titles around.

Chief AI and Strategy Officer at Dialpad. Dan, welcome on, man. Thanks. It’s awesome to be here. Appreciate it. Dan, you got a lot of story here. You got a lot of backstory, and I love to hear everybody’s paths in life. Walk us through, for anybody that doesn’t know you, how did you get into this field, and where did your career start? Yeah, it’s funny. I grew up in the Bay Area, so my first job out of college was actually joining Google. I was at Google for about 10 years. And it’s funny, I joined Google as a couple hundred people. Small startup was probably 250 people in one office. It’s kind of mildly amusing when I look back at being a 22-year-old and kind of seeing that company grow up and become what it is. And I’ll kind of fast forward through some things. Did some different startups. Joined VC as an entrepreneur in residence. Met some technical founders that were working on essentially real-time transcription back in 2016, which was today we know it as closed captioning or transcription. So how do we go take audio and then transcribe it? That became readily apparent in terms of the opportunities to go and power some automation and assistance for sales and support. We did a partnership with a business called Dialpad. We actually did two partnerships as I was leading that business. One was with Dialpad, one was with a business called Talk Desk in the contact center space. And the Dialpad partnership turned into an acquisition. That made a lot of sense for us. And so it was really Dialpad was a cloud communications platform, or is a cloud communications platform. We’ll get into a little bit more. But it was, hey, if we can go power communications on any device anywhere in the world and then use our tech stack to capture and analyze and transcribe those conversations, a whole host of opportunities kind of opened up. And so I’ve been a Dialpad now six years. I oversee our AI product teams, our roadmap, what models do we go and build? How do we talk about this in the market? And so it’s been a really exciting, fun journey.

As the acquired CEO, you kind of… [inaudible] …to maybe beat last kind of a year before you’re kind of shown the door or you go onto something else. But it’s been this wonderful journey of really working on really interesting cutting-edge technologies and what I think are just massive markets that are ripe for disruption with AI.

I love that. I love the journey. I love a good windy path too. You learned so much during that. And I can’t imagine what it’s like growing up at Google when it was that size, right? If you could know now that it was going to be tens or hundreds of thousands, right? How wild is that that you got to have an impact on that? Yeah. Well, fortunately, I think, fortunately, at 22, you don’t realize… I think at that time, I was happy to have a job in tech. And this was back in the O2 tech bus. And so I look back now with a lot more gratitude and perspective than I think at that moment in time. At that moment in time, you know it’s unique, but you don’t know how unique. Oh, for sure. For sure. We’re all brilliant in hindsight.

Okay. So let’s do a little bit of level set, right? For anybody that isn’t familiar with Dialpad, help us understand a little bit about who Dialpad is. Let’s touch on that. And then kind of your role, right? I mean, you’re a busy guy with AI and strategy. So walk us through those. Yeah. So Dialpad’s cloud communication platform. So we power communications on any device anywhere in the world. So whether that’s on your mobile device, providing a soft phone. And then we really focus on building, I would say, specific feature sets for three different personas. The first can be sales organizations. The second can be supporter organizations. And the third is just general collaboration for employees. So I like to say, look, we compete with the Zoom and Ring Centrals of the world. We also compete with the Talk Desk 5.9, Genesis, Nice of the world in terms of a full contact center solution. The difference for us is, look, it’s one piece of software that can do all of those things. The only thing that changes is different features are enabled depending on who you are, right? The persona or the needs that you have. And then our key difference has always been, look, we’ve been invested in AI over the past six years since my acquisition of the business I was running called TalkIQ. I think right now, a lot of businesses, everybody talks to says, look, we’re AI first, AI powered, yada, yada, yada. And I kind of smile and laugh. We are the only business, probably one of the only businesses in all of B2B software that can say, we own our entire AI stack. Meaning we do our own speech recognition in terms of transcription. We do our own NLP models in terms of how do we identify, and NLP is natural language processing for those that are listening. It’s how do you identify things like sentiment? We do our own semantic search. We interact with semantic search engines every day on Google and Netflix. Think of them like recommendation engines, but also they can retrieve information and display it to a user. And then we do built our own large language model called Dialpad GPT. We announced that about two months ago, which was how do we build a smaller parameter model that is use case and domain specific because we think that that can provide a better experience to our users. So then getting into my role is I get to go and talk about all of that stuff, build it, make the decisions of like, what do we build? What do we partner with? Do we partner? Who do we partner? And then it’s a case of trying to be kind of the visionary within Dialpad in terms of these are the features that we should be thinking about that we can leverage these technologies to provide. And then how do we talk about it in the market? So I get to be doing these types of things where you get to be a little bit of like the evangelist. Yeah, I love that. And you know, I think you guys see all this stuff come out, you know, the open AI makes a big, you know, Sam Altman, all these things. And you guys are going, because you’ve been doing this for like, years, this is great that the world has now come around. But you guys are right in the epicenter of being so forward thinking in this and have productized it. And I think that’s what’s, that’s what’s beautiful in this. And I think that’s the goodness in this role is that you’re here to say, Hey, you know, you hear all this nebulous things about AI. And I think people in businesses and customers go, how do I, how do I productize that? And you guys are here with a killer product. And you’ve done that. And I love the evolutions. I love the GPT component. Yeah. I think we have to say GPT on this at least 10 times. Yeah. We can’t say it’s an AI podcast. But no, I love that. I think you’ve just got so much in front of you from an opportunity perspective to pick of, we could steer this thing any which way we want. Yeah, I think it’s been really interesting. You know, a lot of times, you know, talking to an investor, I was meeting with a with a with an investor yesterday. And I think a lot of times, Mike, you said at the beginning, which is like, everyone’s really smart in hindsight.

We have been really fortunate to be like, ahead of this wave and have the wave kind of show up. And I think a lot of times you’re just in startups or just in business in general, which kind of you’re always paddling the catch the wave and kind of adjust. And this is honestly one of the few times in my professional career, I can say like, you’re ahead of the wave and people you have conversations and people would not always kind of get it or appreciate it. And then you’re right at the it’s like, just been perfect for us over the past year. As I said, I think the world was really enamored with open AI and the capabilities of large language models and the opportunities. And we’ve just been like, so well positioned to capture kind of the AI hype cycle on these pieces, but also be, as I said, set up for success where it’s like, we’ve, we’ve got 50 plus people on our AI team, you know, 16 of them are PhDs, which I love to highlight, which is kind of the vanity metric. But I always joke, they’re not the folks that are just graduating school and have taken a couple courses. And that’s great. You want those people that, as I said, you want those employees as well. But these are folks that have 20 or 30 plus years working in NLP and in speech recognition and machine learning. And that’s also an advantage to have that experience. And I was telling a story yesterday at an all hands of the day open AI and chat GPT first became available, we had a code red. And we had an executive meeting late at night talking about like, what are the opportunities? What are the risks? You know, we have this team, are they not needed overnight? And suddenly, you know, what you realize is like, that team is so valuable to us sitting on this proprietary data set and the experience the opportunities, but it’s been a wild year for sure. Yeah, I love it. I love it. And I do think I do think this hype cycle is a little different because there is things to substantiate it, right? There is data, there’s product, there is revenue. It’s different than some of these other hype cycles and bull markets or dot com run ups. And then this is just, there’s so much here. And I do think I completely agree. You guys are positioned as there’s such a perfect timing of it. Yeah, I think that the hype is real. And you know, there’s always the naysayers in the market. And as they said, like, we have conversation, and we’ll probably get into this, you know, like,

I don’t try not to go on too many long rambles, but, you know, there’s the there’s there’s the investors are in on this out there, like, look, these large language models, power chat bots, everything gets automated. So like the voice call goes away. And then the next thing that goes away is like in the emails go away. So everything is just digitally deflected by this smart chat bots. And the one thing I can say is I think that takes a long time to get to if self driving cars are any indication of this is something that we were promised many years ago, and turns out it’s like just really difficult. And there’s so many edge cases. And I think that’s true when you think about the actual workflows and complexity of workflows. And so I really believe that like, the idea that everything gets automated and goes away or can be handled by a chat bot is much more complex and going to take much longer than we think. And so I definitely buy into the hype cycle of, you know, there’s perhaps a little bit of like job, you know, jobs that might get replaced, that are that can be automated. But I think the vast majority of what we do as humans at work in the daily basis is not having AI that would automate but AI that’s there to assist you and help do things. Yeah, yeah. And I think the interesting thing in this and I’ve got a good question for you here on on challenges and difficult things. Like final thought on this, maybe

we want to move at this rapid pace. Yeah. And if we just back up and we’re all beholden to what the customers want to do, and we’re still selling pot lines, we’re, you know, we’re, those are still big opportunities, people are still figuring out what do I do with alarms and what do with backups, right? So you have to look at how long it takes these to evolve. And that’s great for all of us, because we’re going to be helping people modernize for a long, long time. And we’re positioned. And we’re not even in the, it’s like the pregame, you see me laugh. I’m like, we’re in the pregame warm ups of this. And to your point, there’s tremendous opportunity, like there’s features we haven’t even thought about that will show up, let alone the ones that we’re just starting to build, let alone demonstrating like the true ROI. Like I really think you know, what I encourage people is like, understand these technologies, understand how to frame them. And it’s exciting, because I think there’s just going to be immense opportunities. If you’re a seller, like immense opportunities, every piece of software is going to become, it’s not even AI powered, it’s just AI, these technologies get woven into everything. And it’s just going to be immense opportunities. Yeah, let’s, let’s talk about challenges. So you get to see obviously, you know, from a revenue and a deals perspective, what do you feel what what are you hearing that are the challenges from the customer experience, the big challenges? Yeah, I think that the two things is, is, is noise and complexity, right? I think saying and for me, I really, I try my best to avoid saying AI. And what I try to do is really break it down into the specific technologies and how they can help solve things. And so I think one, that’s a challenge for sellers or just customers in the market is like, how, what’s the right thing? How do I talk about this in a way that makes sense to somebody? And I think two is, I think we’re starting to get past the AI hype cycle of, hey, this feature sounds really cool and new, let me use it. And back into the, hey, I understand that. But what’s the what’s the demonstrated ROI? Like, what is the actual impact? And so we’re seeing and hearing that more from customers to say, you know, what are the actual benefits and the value that are provided, as opposed to, oh, that sounds really that, you know, that feat, that AI feature sounds really cool, like, I’ll just give it a whirl. And I hope it kind of works out. So those are the two things I kind of push our own, sell, you know, our own sellers on internally. But anytime I’m chatting, I’m chatting with, you know, with you all in the channel, folks, is that which is like, keep it simple, relate it back to people, and then also make sure you have some ROI and demonstrated facts of impact.

Love it. Let’s talk about, you know, flashback is as far as you want in this. I’m a big proponent. It’s part of what I love in this podcast is I get to hear everybody’s paths, right? So you have struggles, and you work at different companies, and we all have these wild paths. What are some of these lessons, you know, in your role now, look back 510, however long you want, what are some of these lessons that you got, man, these are still valuable, these are still things that we need to know as a company that we need everybody else to know. Yeah, I think, you know, the lessons for me is like, I think the most important thing is taking time to hone your craft is the biggest thing. And anytime there’s new technologies is to really understand how they work. And as I said, they don’t have to be overwhelming. I think this is one of those moments that I tell anybody like, go and educate and read just a primer on AI around large language models and just how they work, and even speech recognition, it’ll help you actually be able to tell that story. So I think like that’s one is like when there’s new technologies, like don’t fear them and be afraid from them and kind of hope that through osmosis, you’ll learn it like

take pride in your profession, take pride in learning, and understand that very few things are going to be overwhelming, like if you spend some time and consistency with them. And then two, I think the other big thing is for me, and I started my career as an SDR, actually, before that I started in customer support. And so what’s to me, the other piece is understanding the importance of storytelling. And I think anybody in their job is ultimately telling stories. And I tell anybody, I’m like, there’s really only two things that matter. And we were talking about being a band, like playing in a band early before we started recording. I’m like,

I equate it to the same thing, which is selling is much like being the front man of or front woman of a band. If you are not passionate about what you are selling and telling a story, nobody else is going to care. If you’re not into the music you are playing, nobody else there is going to care. And I tell people that all the time. So it’s like any time you’re there engaging with a customer, you got to put on the face, you got to bring the energy, like you got to sound passionate. And that’s half of it. And if you can do that, and it took me a long time to figure that out on my own career. So I always tell people like one of those one of the challenges and learnings, I think, is that piece. And then last one is like, you know, and I don’t know, I’m going into like the broad learning and this is what I want. I want lessons learned from being in a band lessons learned from being at Google. And how do you I think this is perfectly valuable. I love storytelling. What do we always say we say stories, stories sell, you know, facts tell nobody cares about the facts, they want to hear the story and why you’re passionate, then that makes them want to be passionate. So no, I’d love to hear this. Yeah, those pieces. And it’s just like, you know, the taking risks stuff, like, as I said, like my career, my career has had plenty of ups and downs, right? Like I think a lot of time, you know, you always as the protagonist in your own story, you’re always like the hero, right? That’s just the way it works. But some I’ve always taken risks. And I think it’s important to live the life that you want. Take the risks and like jobs, opportunities and learnings and all of those things. So I usually encourage people on that side, which is like, take the risk, like good things usually happen. And I have a good quote of like, takes pressure to create diamonds, which I had not heard before. It’s like, make sense. I’m like, nothing easy, you know, the best things are not going to happen through easy pass. Yeah, good, good. All right, let’s talk about let’s talk about what are the customers saying, right? These customers that are out there, you go back, you talk to the customers, we’re going to get into some, you know, talk about deals here in just a second. But maybe talk about, you know, after the sale, what are the customers saying after they picked it? You know, what do they love about the tech? Yeah, for Dialpad, it really comes down to three things. Like one, I would say, we’re in the cloud. So super easy. I smile and always laugh because just, you know, I kind of grew up in the cloud. So having an on premise piece of software is, is unique for me to think about. So I would say like, one is like the ease of, hey, I can go manage this in the cloud. And I think we under appreciate that, which is like, people always ask me like, how quickly can I go deploy Dialpad? Like, well, how quickly can you attach like your can you do single sign in on Gmail, and like matter of minutes, depending on the complexity and route, like the stuff that you want. And I’m like, you can get up and running a matter of minutes. So that’s one is like ease of administration built into cloud. Two is the fact that you can have a single piece of software handle all of your communications internally and externally, sales, support, recruiting. And I think that also can be underappreciated. And I tell a talk a lot of people a lot about like, look, you want your communications stack to power the communications, but also to understand what’s happening in those conversations, because I talk about conversations being kind of the last offline data set. And if you believe in that, and the opportunities of AI to deliver kind of automation assistance and insights, that you want as many conversations happening in a single stack as possible, because that allows you to have universal analytics, you can leverage truly the power of these large language models. So that’s another reason that I would say people pick us is they understand kind of the opportunities around AI and our investments, which is like, look, you can manage this in the cloud, deploy it really easy. It’s going to work for all of our different personas in a single piece of software. That software is completely unified and integrated. So we get one piece of it to kind of single view of analytics across all of these different teams. And then we believe we see that this business is truly on the cutting edge in terms of AI and the innovation and the defensibility of like, we’re going to pick a piece of software that’s going to grow with us, as opposed to something that we’re going to outgrow.

Love it. Here, you know, we always talk about from an engineering perspective, we love when we get in, and we get to understand what people really need. And we find out that there’s a lot more areas that we can help them with, maybe they brought us in for this, and then we find out they have security issues, or they have all the other things, right. And so call some of that out. I want to call that out from your perspective, where maybe give us an example of a surprising win. Where’s a moment where you came in and thought you were just trying to solve for this, and then all of a sudden you gave them so much more than they even thought was possible?

Yeah. And do you want specific customer? I’m just like, should I say specific customer names? Hey, if you can say the customer name? Yeah, I would. So we have, I mean, surprising, and I’m on a customer call late last night in Australia. And the customer stories and the wins that I get pulled into are usually around, as I said, around, if they’re a large enterprise, security compliance, AI stories, and AI, and this is probably relevant, because I’ll talk about this one, because I think it’s relevant to kind of obviously what we’re talking about in the podcast.

The AI piece is, one of the challenges is you have like the innovators within an organization that understand the value of these features, and then you will have kind of not necessarily the naysayers, but you’ve got an I’ll peg kind of the legal compliance folks as the naysayers, where you are, you’re talking about recording, you’re talking about data retention policies, you’re talking about if they’re in a regulated industry, PII, and how to redact that information, and what’s displayed to users. You’re talking about right now, whether that data is pushed to a third party, right, like a chat GBTA, like through open AI, and what are the compliance risks or certifications that they do or do not have. And so these can anytime I get looped in, and it’s the legal compliance folks, there’s a different point of view and a different hat and persona that I have to put on, which is not you get the vibe of like maybe in the jovial kind of like, hey, it’s all like believe in the AI, like all this, like, don’t worry about that. And as I said, there’s a different lens that gets put on. But those are the much more difficult, challenging ones where you’re there to really slow down and say, look, one of the advantages for Dialpad is all of the data stays under one stack. Because we don’t have these other third party relationships, we’re not taking that data and sending it off to a third party that may not have HIPAA compliant or might expose you to this. Being able to talk to them through look, you can set the data retention policies.

Talking them through look, because we do transcription ourselves, and we own the whole stack, you actually don’t need a recording. So if you’re you are concerned around wiretapping and recordings, you can actually get all of the power of our AI without ever having a recorded file. And that’s really unique in our space. Very few people can I don’t even I don’t even think there’s another another player that can do that. And so those are, as I said, like, kind of a surprise, I say like a surprising win for me when we get into the details of just like, here’s like the hurdle, and it’s the legal compliance folks, like they’re going to push on some very, you know, some very nuanced things. But we have some really great positioning. And I think we’re really thoughtful of our approach and actually gives us you know, as I said, I think our team looks forward to those conversations when they can go say like, Oh, we got this covered. And by the way, we, you know, our execs will get involved in a deal and help get this over get this over the line. Yeah, love it. Yeah, legal legal is always exciting. It is I always laughing loud. Yeah, building our general counsel, Joe, Joe favors amazing. And he laughs at me. But I usually joke when we get into product discussions to I’m like the you got to be the visionary person like what are the opportunities and I kind of joke like it’s somebody else’s job to slow us down and have the practical view of like, yes, or good idea, terrible idea. Trust me, I got terrible. Yeah, it’s, you know, the things that you can’t say in those conversations are like, yeah, we should be able to know all the way that conversation is so different when you’re like, yes, we address PCI with ways one to tone of my voice talking through it probably changed like, all right, time to get serious. It’s like time to get serious. Yeah, you know, those pieces. I love it. Let’s, let’s talk about AI. Let’s talk about AI in the channel. I mean, you know, we talked a little bit about product ization. But if Dan has a view to impact AI and product ization in the channel, I mean, one, does a open AI, how do they impact this? But but how could the channel just do better at the product ization of AI over time? Yeah, I think and this is like, this is a biased point of view, right? Like I work at one provider, it’s like one provider of many, right? Of this. If I was one of the things that I love to highlight is understanding that there are many businesses and I’ll tie this back, this can be a little bit of a riddle. Many businesses will say they’re AI powered or AI first, but they’re really a wrapper on an API. And what I mean by that for me, not for people that may not understand API is like, you’re not actually doing anything, you might be a builder, right? If I bring this to construction, but you don’t build your yet, like you don’t harvest your own wood. So you have to go off to somebody else and you hope that they have enough wood, you hope that it’s the right size, you hope that it’s strong enough, all of these things. And that’s when people say, look, they’re an AI first, or an AI powered business, and then they’re using a third party, they’re actually not the builder of anything AI. Their AI engineers, perhaps they’re not actually AI engineers, they’re literally engineers calling to a third party API to build things. That’s fine. I’m not trying to beat the meaning or anything to that nature of it. It’s just the reality. And so there are challenges with that. And that means that that business ultimately doesn’t control their roadmap, they’re dependent on somebody else. And so that can limit their pace of innovation. They probably don’t have pricing power because those third parties can be, they obviously are going to charge a markup to use their technology. So I think for us, like, understanding that, as you’re thinking about the partners that you want to understand, or perhaps, you know, push more or less of whatever it might be, there are nuances in understanding how these businesses are set up, and what they’re actually doing with AI and the teams. And I think that’s like one of the opportunities for people is like, have those types of conversations with the other with the other partners and relationships that you have to really understand, you know, what is this team invested in in doing themselves versus what are they using third parties for? Because I do think it matters. I think it matters around pricing, and giving businesses pricing power. Everyone cares about that. It means you’re going to get into the data retention and the security compliance piece, where if you’re representing somebody that is just using third parties, some of those third parties aren’t HIPAA compliant, or may not be SOC compliant yet. Many of them are startups, or SOC 2, you know, all of these things. And so like, you got to have that question because it’s going, you might bring somebody into the deal and suddenly be really excited about the deal, and then find out that they’re not. And guess what, the legal compliance guy is going to be the blocker on that deal for that reason. And now you’re back to like, you’re back to square one. And so for me, like, it’s, it’s really those pieces where we’re at that moment, there’s a lot of noise, everyone says AI, sit down with your partners, take the time to understand their architecture and what they’re really doing themselves versus a third party, because I think that will help you better position and figure out like, which are going to be the best, which are going to have staying power, like which ones are going to be unique, which ones are going to be the best that you can go position to your customers. And I think they’ll head off future understanding that will head off kind of the future problems or questions that might show up later in the deal cycle, which nobody wants. Yeah. Yeah. I like that. Let’s final couple thoughts here. Let’s, let’s wrap up this thought on advice. You know, as partners listen to this, maybe maybe a partner that’s listening to this has been big into cloud, but didn’t jump into CX, or maybe was in doing a lot of SDWN and network, but hasn’t had jumped into this. And you touched on this a little bit earlier, so maybe just bring this point home for those looking to kind of expand their offerings. Any advice for the individual partners, right? Where would you send them? What would you recommend to kind of get them dialed in and up to speed? Yeah, I think, you know, as I said, the reason I’m here, I think communications in general is the biggest in collaborations. If we just talk about communication, collaboration market, and like people are like, I don’t really go and sell a lot of that or interested in that. I think it’s the biggest team, like it’s the biggest market out there on the planet. So you’re talking about every single business on the planet needs a way to communicate internally, externally. I’m like, that’s an unsolved problem. The next biggest one is like, okay, if you can understand those conversations across all channels, so digital channels, voice channels, messaging, if you can understand them and unlock that, as I talked, again, I was talking about like conversations, so this last offline data set, there are almost infinite possibilities to drive automation, assistance and insights. And we’re at like, again, if you get into like the AI cycle, we’re in like the we’re in the pregame, you know, we’re sitting there taking our pregame warm up shots on that. And so I think truly, if people are thinking about other opportunities, as do you want, like all of that stuff, I would be like, I actually think the communications collaborations, contact center, sales center, right for disruption, massive market, infinite deals, that’s what I would be spending my time on. That’s why, as I said, like, that’s what I’m here spending my time on. I just think there’s an immense opportunity. Yeah, dead on, dead on. All right. Final thoughts. Let’s look at our Miss Cleo crystal balls here. I think we’re in the most rapidly expanding space. You know, besides the fact is looking at Nvidia’s quarterlies last night, I don’t know, I don’t know what they did 84 trillion, right? Maybe they’re our next $10 trillion company. I don’t know. But yeah, obviously doing well right at the epicenter of that too. Where do you see this customer? So much innovation, right? Where do you see this customer experience heading? You can take this anywhere you want, open AI, GPT, productization, anything, final thoughts, where do you just see this going? Yeah, I think so. It’ll give you like three things. So the first, and I’ll talk practically on like communication collaboration. Okay, the first, the first one we got into a little bit of this is like, yes, I think we see the prevalence of much more sophisticated chatbots. And you see that already, like chatbot businesses. And again, I just say this because it’s I say this casually, dime a dozen, I don’t mean that in like a demeaning way at all of that. And just like, we’re gonna see a lot of them. That said, I do think that there is the very real possibility that look, the workflows are much more complex. And that stuff doesn’t get automated. And we get back into needing AI that is going to provide assistance to both the sales teams and the support teams, let alone the back office workers. It’s not about we’re in like the hype cycle of like that all gets automated through like the chatbot. I don’t think that plays out. Or I think it plays out a lot longer than the two years of like everyone needs to go find a job. We’re all kind of doomed. So I think that’s real is like, we’ll see this run up of chatbots, we’ll figure out these workflows are much more complex and harder, it’s harder to get all these integrations, right. And so then AI will shift, I think a little bit back on like the, okay, this is going to be assistance or an assistant here to help me do things, not completely remove the need for me to do it. The second piece is, when I think about kind of the visionary opportunity is every conversation, like this conversation is about data identification and data extraction. And what I mean, it’s like a fancy way of saying, we have topics and questions, you want to make sure that we’ve talked about the topics, and then you likely want to know what my answer is, and perhaps you want to save that answer to something. And that plays out in sales conversations, right? We’ve got a we’ve got a process around like, bands, so budget authority need timing. So we need to identify those topics, have we spoken about it, we want to capture those responses, and we want to write that response back to a system of record like a CRM. So I think like that’s one of the biggest opportunities that’s something we’re working on called AI Playbooks at Dialpad, which is that live agenda tracking, track the agendas, track the responses, capture the responses and write them back. And I think there are amazing opportunities across sales, recruiting, I hate I like interviewing people, I hate taking my notes and writing to a system like you can go help automate and provide assistance on all of that. And then the third one, which is like more general, because just because you were talking about Nvidia earnings on like G and that’s driven by GPUs. And like the largest language models right now need these GPUs to run, because they’re called up, they’re these really large models, based on parameters. And so it’s much like a car to use analogy, you got a really big heavy car, guess what, you need a really big engine have a small car that’s more purpose driven, you need a smaller engine. So I actually think what plays out here too, as well. And this is what we’ve done for our own large language model is we have it running not on GPUs, you have it running on CPUs. And the reason you can get it running on CPUs is you can quickly realize, we don’t need to have this really large language model if we have like these more nuanced, simple tasks. And so I actually see people start to figure this out in the innovators of like, you’ve had that run up of like the GPU, because those were the only models that are available. The innovators and the people that are really invested and understand the technology will realize that you can build these really sophisticated purpose driven smaller models, and put them on less computing resources, which are going to be the CPUs. So I actually think you’ll see more businesses take a similar approach, or understand that approach and be looking for CPU capacity outside of just GPU capacity. Love it. Oh, spicy twist. All right. Good stuff, man. All right, Dan, that man, we covered a lot of good things in there. We could have gone a while on that. Really appreciate you coming on, man. Thanks so much for doing this. Yeah, this is awesome. Thank you. All right, everybody, that wraps us up. I’m your host, Josh,Lupresto SVP of Sales Engineering, Dan O’Connell, Chief AI and Strategy Officer at Dialpad. As always, wherever you’re listening, like, subscribe, you get these notifications right away. Until next time.