Artificial Humans Are Already Here

Hi. This is Nick Earl, the chairman of SI, and you're listening to the IoT and AI Leaders podcast. And my guest this week is Barry Liebert. Now Barry is a serial entrepreneur. He's been involved in over four hundred companies. He describes himself as a unicorn builder. And you'll hear that, you know, at the beginning of the pod, I talk about the fact that he's got the longest LinkedIn profile that I've ever seen. More importantly, in the pod, what we get into is a whole series of discussions about where the world is heading as IoT and AI come together in the world of AIoT. Barry is very heavily involved in a company called HiveMQ, which is absolutely at the leading edge of this. And in the pod that you're about to listen to, we talk about what that's going to work, what it's all gonna mean for companies. What's it gonna mean for the human device interface? What's it gonna mean for jobs going forward? What's it gonna mean for productivity for companies? What's it gonna mean for the ontology of the architecture of the AI system in companies? And what's it gonna mean for the enterprise software market? Both Barry and I have had over forty years of experience in the industry and I just found it very fascinating and intellectually challenging conversation and I hope you enjoy it. So without any further ado, let me hand you over to my discussion with Barry Liebert, who is the unicorn builder and is the chairman CEO of HiveMQ, but has got a lot more experience and involvement in other companies than that. I think you're really gonna enjoy it. Here we go. So welcome everybody to another episode of the IoT and AI Leaders podcast. And in this episode, I'm delighted to welcome Barry Liebert. Now, before I let Barry introduce himself, I just gotta tell you my immediate reaction when You know what it's like, you know, you've got a guest on the pod, first thing you do is you go to LinkedIn, and I've never seen an entry in LinkedIn like this. First of all, a high level description, Barry, Board member chairman and unicorn builder. Thought that's pretty good. Then forty seven different experiences. Normally, you get like six, seven, eight, forty seven different experiences. Been involved in over four hundred companies. So not only do we have Barry on the pod, but we probably have had our first guest who has managed to solve the scientific problem of replicating himself to be in dozens of places at the same time. So with that as an intro, probably the busiest guest I've ever had on the pod, Barry, welcome to the IoT and AI Leaders podcast. Vic, you are far, far too kind. Thank you very much for having me on your pod. Long story short, it's been forty eight and a half years since I left Bald Comey Business School and McKinsey. You know, from very day one, long ago, decades ago, I decided that my calling was to help companies, participate in companies, both startups and large companies, and have been doing this forever. And luckily, I've seen, wow, companies in every industry and almost every continent. I've been privileged to work with some of the best and some of the worst right along the journey, and, lucky to be here on the show today. We probably have only got time in this pod in this pod to talk about one of those companies, which is very relevant to the theme of this pod, which is the convergence and the business benefits and the practical advice of bringing together IoT and AI. So let me just sort of lay the table here. One of the reasons we did the pivot, for five years, it was the IoT Leaders Podcast, and then regular listeners will know, and there are a lot of them, that about two, three months ago we pivoted and we said IoT and AI are leaders. So we've rebranded. And the reason we did that is that we really believe that the AI models and the AI adoption has so far, it's been fantastic, but it's so far been based on training the models around content, which is either written content, audio content, video content. Some people say it's not training, it's stealing, that's another subject we won't get into. The fact is that there by many estimates, there's maybe fifty times more information in the world that's available from things. And if we could capture that and feed that into the models, then you could create whole new business models, unheard of levels of productivity, visibility insight for companies by actually getting IoT data into AI. So net net, the bet that we're making is that the two areas are going to come together and IoT is going to be turbocharged as a result of merging with AI in particular information from things being fed into the AI models or the other way around. So Barry, when I looked at HiveMQ, one of your companies, it seems to me that that was right on the money, right a bull's eye in terms of the premise. Have I got that right? You do. Spring of last year, one of the batches I worked for, which is called Mobadala, called me and said, there, I got another company for you. This is my standard phone call from an investor. Can you look at this company for me? And I looked at and this is in March or April last year, I looked at, I said, gee, I've seen these companies before. I was an advisor on a company called WebEx and Super IR. I'd been an advisor to AT and T back in the nineties. I had seen data streaming companies before. And so that something I recognized. I had not been in a data streaming company between devices as opposed to between humans. And I was already in the AI business since really machine learning since two thousand eleven and a bunch of other companies since then, like Anaconda since two thousand seventeen, eighteen, and nineteen. Not in the not in the model portion, financial model portion, but just in its application. And I thought, wow, this is a big opportunity. A long data streaming, but between people, it was between devices. And there was plenty of research coming out from my alumni from McKinsey that was saying, so you're be called AIoT, the convergence between models and devices. Now one might argue, you might argue, anybody might argue that devices are limited to physical things. I'll just use this as a device for a second. That phone is a device and two devices. I don't mean two phones, but two devices on a factory floor or two devices that can be connected, which they are by people like us. The information between them is streamed. Models can be built both at the edge and in the cloud, and they'll those models will allow both the machines or devices at the edge to operate more effectively, more efficiently, and will adjust themselves in time. Or I'll give you a big idea. You and I are already at IoT. What do I mean by that? You and I are a device already. And what do I mean by that? There's research that says these are just exoskeletons. We the research says that we would rather have our arms literally hand amputated than loose and leave these devices. That's how important to our kids. And so you might argue we are a device now. We're these devices and we are device extensions. And so robotics are part of the IoT world. And, you know, there are companies called NeuralNet, which is people like Elon Musk working by putting devices in our means. I don't know that there's a from a profound perspective, I don't know that there is not a convergence between the human and let's just call nonhuman. You know? I don't mean artificial humans. Just nonhuman devices in the world that create an explosive opportunity for IoT and AI to converge as devices, machines, and data come together. That's an interesting premise. Funny enough, I was listening to there's a Sunday Times newspaper. They do a good great pod. And they were talking in the episode that I just listened to this morning of this whole concept of the blending, the merging, the blurring of the human device interface. And they talked about Neuralink and they had a bit of a debate about whether or not you'd want to go first to have the wafer put in your brain or you'd want to wait until it was sort of something that could be attached to the outside of your skull. But the point being is that knowledge information processes are gonna be split across a combination of the device and the human. And the device is just a gateway to the cloud and all of the information. And so it sounds completely science fiction, as you point out, that interface is blurring now. And it's not just Neuralink, Elon's doing a lot of experimentation in this area. He's also with his robots and his predictions of what's he gonna ramp to like a million a year or whatever. As work increasingly, cognitive work and physical work increasingly gets automated, the role of humans has to move up the stack. And in some way that where the connection between the two has to be more than just looking at the prompt output and saying, I don't believe you. Can I double check your answer? And I think that's what you're sort of saying is that the big vision is that everything starts to blur. Humans become devices. That's the first time I've ever heard anybody say that as as as we are extensions of our own devices. It sounds a bit dystopian. But if you think about Taylorism, William Taylor, that he wrote long, twenty years ago, that really humans are artifacts of completing a task. You might argue machines are the same thing. Right? And one might argue, I don't think it's true, that we are somehow separated. This is a long conversation from our environment. I don't think that's true either. Right? Or somehow we're separated from each other. I don't think that's true either. I think what you're saying is that more and more our environments, our technological environments, our informational environments, those things are merging, and critical to those things is data streaming. So you and I are having having conversation here. Now we're doing voice data streaming. You can we're recording it, we can send it out and have an LLM look at it. We can have it transcribed. We can have it ultimately synthesized, and then it can be reconnected into something that can be used for further intelligence. I think, you know, my entire life, I've seen the continued for forty and a half years, continued technological advancements change the way, I hate to say it, the way businesses are actually going to operate in the way many companies think they operate, which is they still think they're somehow operating in some prior periods. And the fact that it says IoT and AI, it might be better than AIoT. Could I suggest that they're separate? Right. They're not separate. They're Yes. And they're coming together in a at a speed I've never seen. I bet you said something for you. This is this large language model, which is training on all the words of history is changing the way. I'm amazed at the speed by which my companies are doing things that I never possibly imagined. It's only humans could do it. And that's not really true anymore. I believe the first foundational level of that is data streaming. We have to stream the data, not just between you and me, but between devices and devices in the cloud and devices on the edge, and then take the insights that allow the devices to operate more efficiently and autonomously. Those are massive opportunities, as you said at the beginning of this pod, that says the amount of data that's going to accrue to the large language model providers and to slow language modifiers and to edge providers and cloud providers is gonna be exponential. When we first rebranded, you're right, we did think about AIoT, but I think somebody had already grabbed it. So we had to call it IoT and AI. But when we first rebranded, we had a guy called Rob Tiffany, who's based on the East Coast of the US, a friend of mine, he's been around also for a long time as you and I both have in the industry. And he's now research director. He's been at Ericsson, he's been at a bunch of companies. He's now research director at IDC. And he was talking about this in, I think it was like three or four episodes ago as what we're able to now create as a new enterprise brain. And the way he described it is that if you could gather fifty times more data than you can gather today and feed it into a model, and you talk about streaming, and you could get an almost real time feed, you could actually create a dashboard for a company that is something unlike you've ever been able to get before. I remember in the nineties when SAP introduced R3 with the Uber platform for enterprise software, that you could actually define, you know, horizontal workflow. And that of course created business enabled business process engineering, and it was radical and companies productivity improved, but it costs a lot of money for an SAP license. And now what we're talking about with the democratization of AI and the cost of everything coming down so rapidly, the idea of an enterprise brain is like this super dashboard where you can see everything that's, if you're running a factory and there's thousands of devices or multiple factories, the ability to get all of that data fed in, maybe through edge aggregation or whatever it is, right down to the individual components in your machinery, and then have a dashboard allowing you to make real time decisions or allowing the AI to make real time decisions based on preemptive actions in order to optimize output or reduce downtime is something that has never been possible before. And I think that's what you're saying is that we're heading towards that capability of having an explosion of data at our fingertips. And therefore the companies that take advantage of that will have tremendous opportunities for productivity and capability to create new services under business agility. If they can harness all that together into whether you call it an enterprise brain or an enterprise dashboard. But it's something that we've never had before, but it looks like it's now becoming possible. I think so one step further, but absolutely. I think those machines will ultimately act autonomously the way you and I act autonomously. Those machines will receive signals, which means both downstream and upstream. They'll they'll send signals to other devices, which they knew already. They'll send signal to the cloud, which they can do already, but those signals will come back to them and those machines will adjust their behavior. So it'd be heat, sensors, activities, flow rates on chemistry, flow rates on water, flow rates on productivity. They'll also go one step further. They will send signals to the supply chain and say, I'm about to go down. Send me some send to my my repair team, my maintenance team, a new particular part that I need Before I before before it breaks. Yeah. Before it breaks. Right? And so all those things are gonna happen. And and it for me, it's not there for, you know, a reality. Like, I get all those signals from my Teslas, right, as I drive around. Signals of things that I need to do, where I need to go, where I need fuel, how hot I'm running. It even tells you when it's breathing nowadays, which is incredible because I've been up to be in the mountains right now, so tell if it's running cold or hot. All those things are gonna happen, and they're already happening. And so I think we at HydroQ already have something called Pulse, which allows the things from Edge to basically transmit that data and go back out to them under something called unified namespace. But we are building an ontology no different than something like other some of our major competitors are doing, which is Alex Karpas figured out. Right? And and realize that you can build these capabilities to provide complete operational intelligence that are autonomous actions and will provide feedback back down that. An example of that, since the first of the year without using an ERP system, we had high BenQ built with ClawdCode. Everything you just said that is at the edge, which means we're a hundred fifty people firm, we can reset everything from the devices to the people, the communications, to the financial activities, to sales and marketing, to things like Slack channels, to what you and I are doing today in a single environment. It's beyond my effort and has been in my I'm a McKinsey guy. No. I mean, now you you'd never imagined me. Mean, that was unheard of just a few years ago. I mean, it was Unburdled. As a McKinsey guy, I remember being I'm gonna exaggerate it, but you get the point. Forty years ago getting paid by the page, I would say the bigger the report, the more we get paid. Yeah. I was on the other end of that. Go. And I worked for Hassel Podner at SAP twenty, thirty years ago when he first started trying to explain to him that I think it would all become democratized, and these big extensive implementations wouldn't work. And so they have worked too successfully. But that when you have Claude Coed that's rewriting COBOL, if you saw that announcement this week when I Yeah. I did. Yeah. Did. Yeah. Yeah. I mean, the the the the the, you know, Moore's Law one point four times a year and and the way that in particular areas like the latest release applaud and the programming capability, and depends on which benchmark you look at, but it's like 4x to 6x to 10x per year. You know, the point that you're raising Okay. So let me ask you what's gonna happen on two different fronts. The first one is we already see it happening in the marketplace, which is the collapse of the share price of the big software companies. I also know Hasso, I used to be based in Germany when I was HP, I was the account exec for SAP. I used to call on him in Germany when Waldorf was just a small office with cows outside the window. And I also, when I was based in Silicon Valley, used to meet with him. And the biggest problem was to stop him ending the meeting quickly if the speed picked up in San Francisco Bay, because he wanted to go, he didn't wanna talk to me. He wanted to go off and get on his boat. But the fact is that over the last, I would say six weeks, the enterprise software, what you just exactly what you just said, the value of enterprise software companies like ServiceNow, that was the leader in agents and you know what, they're down forty, fifty percent, SAP's down, PeopleSoft is down. Because people can do what you've just said. I mean, they can kind of create this enterprise layer, this really quickly, really cheaply. And so the enterprise software share prices have dropped hugely. That's the first of the two questions. The second one's going to be around jobs and the role of humans, but I'll come back to that. Do you believe that that trend is going to continue? Are we witnessing just as SaaS killed the enterprise license business? Are we witnessing the ability to create these capabilities, killing the the SaaS enterprise SaaS marketplace? Yeah. I've been running about this for a year or two now. I do think everything that comes after what it came before begins to deprecate or degrade the prior business calls and Yeah. It just does. Right? So, you know, you think about all the things that have the last forty years, all bricks and sticks, businesses, all the Walmart's doing a bit well. Basically, you do well because it's ecommerce engine. Amazon was the next version of that. Now Amazon was just declared the largest company in North America, which Walmart once held. K. And so new technologies come along that deprecate the prior technological business models. Right? You see the recent bankruptcy of some of companies I work for from the human market to Pertodf Goodman. The way you get those luxurious products today, I would say. Luxury today is getting it at my doorstep the minute I want it, which is nowadays Amazon. Delivers at the same moment. Literally, the same day is sort of crazy to me that, you know, next day was fine. All of a sudden, it's the same day. So Yeah. These new technologies change it. And that's what happened between cloud and prem. Right? I'm I do think that's just gonna happen now between SaaS and AI. Now it doesn't mean that the current valuations of SaaS companies are gonna stay there. They're in a battleground right now with AI people for their customers, and companies are trying to decide how to spend their money. If you look at today's research, number one in every CEO's agenda is spending money on AI. It's to fundamentally take advantage of these new tools and to reduce their cost of operations while they accelerate their productivity and revenues. So now the SaaS companies have a problem. Just like the on prem companies do, they either have to change their entire business model to what's called health space, right, And usage based, SaaS based, which is perceived less. And they have to embed these new technologies while still keeping their current revenues. Very hard trick to do to fly while you're fueling at the same time. And these new AI companies, which have consumed huge amounts of capital, that seventy six percent of all new investor capital, all investor capital, public and private, is going to AI companies. So not only SaaS companies can't raise capital, they have troubles getting capital from customers as well. So the question is there's a battleground between SaaS and AI for who's gonna win the battle of the investors and customers. And one step further, who's gonna win the battle for human talent, which is where you're gonna go. Before I go to the human element, I I share a a personal experience because, you know, there's as the Italians say, you know, nothing new under the sun. It's a pretty common Italian phrase. I, when I left Silicon Valley, I went to work for a REBA in two thousand and one, and it was an independent company and that was the hot company. I remember it was worth forty four billion dollars with seventeen hundred employees and it was the darling of the NASDAQ. And then what happened was, now to be fair, coincident with the crash, but what happened was they were selling licenses. So I used to run all sales and operations outside of North America for them, And they were selling licenses on average of like seven million bucks a license for Ariba. You know, you pool your procurement with other companies. And so instead of buying your own shares, discounting them, negotiating a purchase agreement, you get twenty companies buying the same shares and you get the value of aggregated spend. And they took a quarter of a percent, very simple business model. And then what happened is SAS came in and they pivoted from license to SAS and told the stock market. And I was there in the room, know, well, don't worry about it because over a three year period, we collect more money and it's much stickier. And basically the market said, yeah, great. Okay, well, we'll just leave and we'll come back when you've proved it. And then the crash happened and the share price went down ninety eight percent. They held the record for the biggest drop in the share price for a company that survived and ultimately they were acquired. So if we're gonna see that again, because the enterprise software companies, and you talked about McKinsey, when I was at both HP and Cisco, the minimum price for McKinsey for a report, and it was by the page, it was like half a million dollars for a report. And it was created by very clever graduates in India. I mean, would arrive the next morning with a half million dollar invoice. So if that's gonna happen, then yeah, then something else is gonna rise up and take its place and it's pretty clear what it's going to be. But what about the role? Let's go back to the humans. We talked about the really futuristic, the blending of the human device interface, but let's put that on the one day pile, but let's go to the today pile. What is the role of humans in this world? You know, there's a bunch of benchmarks, the latest ChatGPT five point two cognitive work benchmark, whatever I saw, it says that it can do seventy one percent of cognitive work. We know that something like only four or five humans can beat AI on math, on the fields metal maths tests. Coding is, you mentioned Claude, coding is now, it's gone. And it used to be send your kids to university to do STEM science, technology, engineering, mathematics. Now they're the very subjects that are being eaten within months by AI. So the big debate is everywhere, and I'd love your view on it. So what's the future for the younger generation who are perhaps leaving university with debt and are suddenly struggling. And I saw one last stat from Peter Diamandis this morning on his moonshot pod saying that graduate unemployed, graduates in North America are now the number one cohort for the people who are waiting the longest to get employed. So it's just rising. They're coming out of university with skills that are competing with AI. So before we get to human device blending, how are we going to look after this next generation who can't get a job? Yeah. So I think because I'm an optimist. I wouldn't keep doing this if I wasn't. I'm an optimist about about humans. I'll come back to AI in a second. About humans. All my wife, you know, has seen these types of debates about technology in humans. You know, humans are not gonna have jobs and stuff like that. And generally, over the last forty eight years, there has been a job for a human, at least in North America. You know, we're, like, under five percent. We're less than that. We're under three percent unemployed, four percent unemployed. So I'm a big optimist. This is technology advances human capacity. AI has something different, though. AI fundamentally forces humans to reconsider what is great about them and what can they do that's truly special. And so I think for the knowledge worker, looking over the physical worker. Yeah. Because that's a robotics question. We'll have to come back. Yes. That's a different question, on manufacturing. Yeah. For another meeting oh, That's another part. Right. That's correct. So for I'll call it the knowledge worker, if you don't mind. The university, the US students coming up, the knowledge workers. I think there's still a huge opportunity for them. The way they have to think of their jobs is about what I'm gonna call framing. So what McKinsey taught me a long time ago, my brother was from PCG, my brother myself was from pain, was that it's the questions you ask, not the answers you give. And so getting the right questions aired helps, I think, humans, they all win. And I don't mean win versus air. I'm not suggesting it's a win for their lives, which is how do they think about their lives, their companies, their opportunities? If they ask the wrong questions, which is how do I get a job? I don't think that's a winning question. I don't think that was ever a big question. I feel excited cynical. He says start with why. So I think That's right. See, never really been challenged with what is my why? Why am I here? Now I have a why. It's to pay it forward, believe it or not. Having been lucky enough to win along the way, I care about paying it forward to all these companies and helping those young professionals, new druggolic companies, fundamentally develop themselves inside the world of AI that we live in. And so that's my calling, and it's been my calling for to help people. And so I think if anybody, whether they come out of university or they run a company, they have to now say, look. Artificial humans, I'm gonna use a felony expression than an agent than AI. Artificial humans are now here working with robots. They can do things faster, better, cheaper. They don't have they don't need coffee breaks. They don't need to sleep at night. They And they can joke to each other. They can learn from each other. Yeah. They can now get each other. They're building their own religions. If you're watching what's going on. There are a lot of things these things are gonna these artificial humans are gonna do, including taking some of our resources, water, energy, electricity Yeah. Natural resources. They're gonna consume just like us natural resources. These natural these artificial humans are working side by side with us. So think of it as like another species in the So how do we work with them? How do we, a company, a leadership team, a board, leverage them in a way that they do things that we they can do better than we can do, but there are things that we can do better than we And vice versa. That's correct. Or think about it, following companies. We've gone into some interesting areas on this part, haven't we? I didn't think we would at the beginning, but that is the idea of there's another species already here. I always like to use examples. There's a big debate here in the UK, which has been, of course, in the US to do with immigrants coming here from other countries. And, you know, we've got this whole thing about the English Channel and stopping the boats and you had the wall on your southern border and whatever. But there was again, another debate I saw the other day, I can't remember who it was who said it, said, you know what, we should really be, that's not the thing that's really going to be our biggest problem. The biggest problem is no one's saying about the million, ten million agents that are gonna arrive and take our jobs. That we actually think of it as, oh, well, these people who are fleeing war zones in Africa are coming to the UK, but no one's saying the problem is there's gonna be a million agents. And on that point, yeah, every wave of technology, this is the glass half full, where every wave of technology has always created new jobs. Of course, this one is, we've never seen cognitive work done and with the robots, different pod, physical work, but there's two aspects of that, which are different. One is there will be a gap between the adoption of the technology and when the new jobs appear. And I think that's what we're looking at right now. That's the graduate issue because the graduates are coming out now, but we haven't worked out what the new jobs is yet. So there's like a valley of death. So the question is how do we get the next generation? You talk about paying it forward. How do we get that next generation across that gap? But the bigger question I was gonna ask you is that the future is already here. And as somebody famously said, but it's just unevenly distributed because if you take something like call centers, so AI software is absolutely super, well, it is so much better now than the initial chatbots. You can call an AI call center at a major company and you're really hard put to work out. It's not a human who's talking to you. They have empathy. They sought access to information. But what that's doing is it's enhancing the customer experience in say the UK or Europe and the US, but it's gonna hollow out certain countries. Because there are countries like the Philippines and large swathes of India. There was a big AI summit this week in India where there are millions and millions of people who are doing those jobs today. And those jobs are disappearing really, really quickly. So do you think that, do you think there will be national winners and losers because of the way AI is being adopted. That some of these jobs that we outsource to lower wage countries will be the first to get automated. And therefore we're kind of almost in danger of increasing the divide between what we sometimes call the developed world and the emerging world, because in the emerging world, we used augmented, we use labor, lower cost labor. We put our manufacturing in China. We put our call centers in the Philippines, but actually that's the type of work that's gonna get automated very, very quickly. So it's not gonna affect every country the same way. And every company, right? The same thing. I think you're gonna see, know, I think about these sourcing models as simple. Insource, labor, and source. Like outsource, we did outsource to low cost labor right around the world. I don't mean just North America, but we outsource to the world and so did England. If I remember correctly, I don't I don't if you're thinking I didn't give out service to labor pools around the world because labor way So it's pretty easy. Yeah. We we didn't get our PR right. We didn't or our actions. We didn't. We didn't. We did. You did Different different time, different implementation model, one that we try not to mention anymore. The western operator's period. I know. Right. Nothing is new, as we just said, Then we crowdsource, which we we basically allowed anybody to you know, there's a big large crowdsource movement. So network is crowdsourcing where we give all our stuff to Facebook and Twitter. We crap companies crowdsource to us, whether it be content or photos. And now we're just a source AI sourcing rather. So I'm assuming we'll do robot sourcing as well. We basically have done these things before. Now what I agree with you is and I don't have the answer to this question. It I've never seen it this fast. Never ever. Yeah. So normally it was a generation. It took centuries for while, but it took generations. Took decades. Now it takes not even years. It takes months. This stuff's going to It takes months. And that was my point about the gap. The The adoption is quicker than us trying to work out what the new jobs are. That's the valley, I think, the real issue is will will there be new jobs for humans to demonstrate those soft skills like critical thinking, judgment? Yes, they will. But how long will it take us to work out what those jobs are whilst in the meantime, we're laying off people and we're not Correct. So I do agree with you because of the speed by which this takes place. I'm still an optimist. Right? That said it's the leaders of countries and companies that can can but understand the speed and not reject this speed will win. And that means everybody can win if they really unders Like, I look at HiveMQ, which we started with. You know, this was a data streaming company. They did not believe that AIoT was emerged opportunity. They didn't understand this. Right? Go back to March of last year. The founders were saying it was just IoT, not AI plus something. I'm like, that's impossible. I've done this before. Was at a Ford company called fixed software, which we didn't saw this back then in twenty one twenty one when we sold to Rock. We saw the connections coming between these exact same things, devices, data, data streaming, and intelligence. So I think that there is and and our customers at High Dung, you were asking for this, which is, can you help us understand the ontology? You'd call the enterprise brain, but the ontology, which is all devices, all the movement of all the information, all the human interactors, operators. Can you help us visualize that in real time? Can you help us see that as an ontology? Always those are semantograph or knowledge graph, lots of words for it. People use. Can you help us see all that? And once we see it all, you know, historically, we probably would have cost ten, twenty, thirty million dollars for McKinsey. Wait. They still couldn't have done it anyway because by the time it was done, machine would have moved. The people would have relocated. The data would have Can you help us see that in real time? You know the next question they're gonna ask us at HybemQ is, can you help us begin to automate that and build an what you call an enterprise brain or an ontological brain? And then they're gonna ask us, it's coming. Can you help us build the sensors that go back and allow the machines and the human operators, right, the artificial humans and the real adjust their behaviors based on the cognitive intelligence of it all. That's going to happen, and it's happening already from their place. I totally I totally agree with you. I think that is how this problem will be solved. That's kind of what I was pushing on. I'll use one last example before we get to the end. About ten episodes ago, we had a guy from Turkey actually who had built something sort of similar for healthcare. Healthcare is absorbing so much money as we all know worldwide and it's really inefficient, particularly in the UK, the National Health Service. It might be free but it's really inefficient. Basically what he did was he just connected, he put sensors on all the things in the hospital. He put a sort of wristband on every person who went into the hospital. It was the same model of wristband that you get when you go to a Coldplay concert, by the way. He was quite proud of that. And then he put a device in the pocket of every nurse, every physician, every doctor, whatever. And so he was just basically tracking patients, people and things, but then he did exactly what you just said that his company built this ontology of what, and it wasn't an ontology by department, the radiography department, the accident and emergency department, the pharmacy, the operating. It was an ontology by process, like a hip operation. What does a hip operation really make up and where are the gaps? And so suddenly you have this dashboard of processes. And then if you're a hospital manager or a clinic manager, and especially if you're managing multiple, which often I do, you could look at the efficiency of various processes across your different healthcare entities and actually say, where is the problem? The problem is actually, it's not visible. It's people go to accident emergency. They get triaged quickly. They get treated fairly quickly. But actually a large part of the reason they're there for hours is they're waiting for the medication to come out of the pharmacy to take home. And that's not visible unless you have this ontology or this sort of new layer. And he was using AI and then he was creating agents to optimize the processes. And I think there are a few companies and he's doing it mainly in the Middle East, in Middle East hospitals and in Germany, I believe. And I think that is the answer. I think that what we have to do is we have to build this ontology and we have to collect as much data as possible. And out of that will come a tremendous opportunity to do new things. And they will involve humans working collaboratively with the software and with the devices. And I think that is the hope that, you know, we need to get there because I still waving the red flag of what's gonna happen to a generation of how long is it gonna take us to get there? But I think that is why it is AIoT, that that is actually the world that will become commonplace going forward, whether you call it enterprise brain or the ontology, that is the way companies will run-in the future before we blend our brain with the device. Correct. Every company has a good wow. I need to be a data streaming company. Why do I start there? This is why I take AT and T and Webex is what I but I when I came to this company, I was like, I've seen this before. Now both those companies, Webex and and and AT and T, streamed a lot of data between people. I associated Subaru IR, which is the founder of Webex before he sold his company. What look. You have this massive hose just like Zoom. Massive hose. Water flows through it all day long. What do we know about the users? What do we know about the devices? What do we know about what happens as a result result of those conversations? He would say nothing. And I put AT and T parties to something. That's not good news. That's the light broadcast. It's not good news. Right? Yeah. And so and you can see what happened with both those companies. They've become, you know, personally, AT and T or Baker. People may not remember that. And Southern Bill bought those assets. That's right. There's a big change here. People not realizing. Look. I'm an optimist and especially at HydeMQ. I'm an optimist because I think, gee, we start with data streaming. Our customers already asked them for intelligence, which is what they need. They're gonna ask the extra operational activities and insight called ontologies. And then you know they're gonna ask for automation. I don't mean automation. Automation the way you and I think. Automation, meaning let my devices, my people go. This is a funny old expression. Let white people, my devices go. That's correct. But let people go, let white devices go do their thing and let them and get rid of all those inefficiencies that you just talked about in the process and be able to observe it. Observability and governance, not the way you and I think about it now, but observability and governance will be critical in that entire process. So, you know, you can see, I've been optimist, I've done this for forty eight years. It's a great it's a great point to finish on. We've covered Barry, we have covered a lot of ground, and we've gone into a whole series of areas, and there's even two big areas we touched on, but didn't even cover was what's gonna happen in the physical world with the robots and what's gonna happen in terms of the blending of the device with the human, is, wow, there's so much more here. But listen, why we wrap it up? I really enjoyed this. I hope you did as well. I hope our listeners did. This is exactly what we wanted to do with the new format of the pod is to actually talk around what's happening, get a global view and get a view from visionaries who see what's happening and are at the cutting edge. Your work with HiveMQ, I think is really, really relevant to this. So I wanted just to thank you for being on the pod. And I want to encourage our listeners and viewers who watch it to actually check out HiveMQ, because it is a very, very interesting company that I think, if I'm right, it actually has a European footprint, doesn't it? Isn't it? Engineers in Germany? Got that. We found it in basically around Munich. It's Munich suburb. Right. Recently, and we have we're opening up in offices in Silicon Valley now. In Silicon Valley. So you're going from Europe to the US. We don't hear that very often over here. We always we we think we're we think it's all coming the the other direction. Barry, really enjoyed it. Thanks very much, and thanks for being a guest on the pod. Thanks for helping me tell you. I appreciate it so much.