[Music] Welcome to another episode of Tech Unhinched, where technology means human. I’m your host Rabia Chave and today we are joined by Rich Thrani, the CEO of TMC and the mastermind behind the IT Expo Tech Super Show. Rich has been a powerhouse in the world of cyber security, communications, and IoT for decades. And he’s not just a visionary, he’s someone who’s actively shaping the future of these industries. Right now, he’s also an investment banker at Four Points Capital Partners, diving deep into the world of mergers and acquisitions. Not only this, but he has also earned recognition from the likes of Forbes, The Economist, and the New York Times with a wealth of experience that continues to influence the tech landscape. We are super excited to have you today with us, Rich. Thank you so much. Thanks for having me. I’m really happy to be here. Rich, you’ve been a tech visionary for over five decades, but outside of the world of cyber security and M&A, what surprising passion or hobby you’re into that might surprise our audience today? Well, I don’t know if it’s surprising, but I I try to be as fit as I can. I try to work out. I try to hike as often as possible. And uh I also like music. Those are my my basic hobbies. I I try not to do too many things that take away from my ability to work cuz I’ve got a busy schedule and I I like to devote as much time as possible to getting things done. The other day I found out that the CEO of Goldman Sachs is a DJ and this is something that that he does for out of his passion and he has an entire page there and you know all his music and I was super impressed. You know a good thing about music is that it doesn’t necessarily interfere if you’re working as well, right? You could I like to have a lot of music when I’m working. It’s a great hobby. You know, typically like the upbeat music uh normally when I’m working, but then there’s sometimes depending on what I’m working on, if it’s deep thought, then I’ll change it to a more relaxing kind of music. But I love music and I’m also uh listening to more and more podcasts these days. Fantastic like like this this kind of podcast. So well Rich, you know, let’s jump off a bit into the topic warm-up and context setting for it. With AI agents increasingly managing critical systems like network security and communications. Do you think this is the logical next step for tech or are we at a risk of letting automation run ahead of our ability to manage it? Well, you know, the challenge we have is that we’ve got two sides to technology. We’ve got the good the white hats and the black hats. And so we’ve got hackers using AI right now on there’s an increasing number of attacks with AI and enhanced by AI. And so to defend ourselves, we’ve got no choice but to use it as well. And so our ability to manage it is almost mandatory that we have to figure that out. And I think that’s where we are as an industry trying to figure out how these different AI agents are going to help uh run our companies and and keep us safe. Yeah. No, that pretty much sums it up. But you know, with your background at Apex Technology Services, which provides cyber security to Fortune 200 companies, how do you view the surge of AIdriven security tools which are now everywhere? So, yeah, Apex is a it’s an MSP, a managed service provider. like you said, we’re we’re providing um IT support and cyber security to companies from the Fortune 2000 on down that because we have that experience that that company is constantly looking for the best tools to support customers to keep customers secure. As an MSP, you’ve got no choice but to focus on really securing your customers because if your customer has a has a problem, they could be out of business and uh you want to keep your customers happy and and running. And so the AI tools are constantly evolving and the challenge is trying to find the right tools for the right clients. Not being too far ahead of the curve, but far enough ahead that like I said that you’re defending against these attackers who are getting more sophisticated. So it’s a continuous learning curve and I’d say in the IT service business we haven’t seen such a learning curve uh and need to advance since the change of the industry from just making your money by selling hardware to manage services where you would start to make monthly recurring revenue. So it’s a it’s a new it’s a paradigm. It’s a disruption. It’s a change and every IT service provider MSP has to get behind it and and stay on the leading edge. So Rich, in your experience, you know, where a lot of companies are rushing to adopt AI agents for cyber security automation, where do you think the enterprises are getting the AI adoption wrong? You know, is it the strategy, the execution or the deeper issues like data integrity and organizational culture? Well, we’re early on and AI chatbot agents were maybe one of the first places that companies were experimenting and and what we found is that, you know, the AI tech itself is not perfect because it hallucinates and uh dealing with the hallucinations has been probably the biggest problem that companies have had to deal with in the last year plus where courts have determined that if a company’s uh chatbot says something, you have to stand behind it. So if it makes a mistake, if there’s a hallucination, you have to figure out how to address the potential damage. So how do you minimize those hallucinations? I think the challenge we have as an industry is deploying AI that is not doing things that it it should not do. And and I think that’s where we are right now. We haven’t gotten there yet. I mean there there are things you can think about like you know if you were to run your answers uh from a chatbot through a few different AI models you know theoretically you’ll filter out all the hallucinations and there are things that we’re going to see coming to the industry going forward where that’ll be more commonplace but right now that isn’t happening and so we do have challenges and hallucinations probably the biggest that I’ve seen but also you know u rich you’ve been a trusted adviser on you know merger and acquisition deals for years and have been watching tech companies go through uh transformative acquisitions in today’s climate. How should investors weigh the real value of AI when evaluating companies for acquisition? So just because I’m an investment banker, I have to be careful not to give investment advice. I’m just have to make that disclaimer that I’m I I’m not giving investment advice. But having said that, I was listening to a podcast this morning actually, uh, Mark Andre of Andre Horowitz, who I’d say is probably one of the most thoughtful investors, intelligent investors out there. Certainly, one of the most experienced, successful, I mean, just can’t say enough good things about him. Whenever I listen to him, I just feel like I’m learning infinite amounts of information or infinite amounts of knowledge. I’m just getting smarter. this morning he was saying that he’s pretty much all in on on investing in companies that are AI first looking to disrupt traditional companies and he thinks that you know in a lot of cases new AI companies are just going to wipe out the existing ones and I think that’s the case and so the challenge that companies have we’re early right now so I don’t expect an existing company to be AI first right I don’t expect General Electric to wake up tomorrow and say oh we’re AI first, right? That’s a huge example. We are all now starting to take baby steps and figure out how this stuff works. That’s the first thing we need to do. But then the second thing we need to do is rearchitect our organizations as AI first organizations. And that is the more complicated thing that needs to happen. Uh but keep in mind that there are startups with a lot of funding behind them from Andre Horowits and others. Their website’s A16Z. I mean, it’s not a plug for them. But just in case anyone wants to learn more about them, there are a lot of uh VCs putting money in startup to wipe out existing companies. So if you’re not going to be AI first and another company comes in and is able to produce similar quality work with a fraction of the expense because they’re using agents instead of humans, they’re going to undercut you on price and you eventually are going to be out of business because you’re you’re just margins. You just can’t you can’t make money and so you’ll be wiped out. So ultimately that’s where things are going. And so I think that to me is the most exciting thing is is trying to figure that out. But you know it’s early days you have to kind of see what’s the company culture like. Do you think this company culture is adaptable? Is it changeable? Changing company culture is not easy. So is it a technical culture? Is it are there scientists behind it? Are there techies um engineers? It’s I mean that’s what I’d be looking for because if it’s a traditional you know bricks and mortar business with paper and they haven’t really automated it’s going to be really tough for them to leap to AI first. So I’ I’d be cautious of that. So you know let’s uh reflect a bit on the AI in practice and what’s working out there for the companies just like Symbian they are using AI agents to manage security operation centers helping automate threat responses. So have you in your experience seen any examples where AI has dramatically improved efficiency and reduced human error in a security operations center. The probably the biggest challenge after dealing with actual real threats is dealing with the hypothetical threats or the theoretical threats or the we call false positives. AI is is drastically reducing the number of false positives that humans have to look through and it’s been doing it for a while. Without AI, what you end up with is security researchers or security operations personnel flooded with thousands or tens of thousands of of false positives. And trying to find a real hacker when there are a lot of false positives is the challenge that I think that most of these operation centers face and the personnel face. So the sooner we can minimize the false positives, the better. And AI is doing that right now and it’s getting better at it. And thank God because again there’s more attacks. there’s typically less money to spend on on personnel in these centers. So AI has to has to keep playing an important part there. Yeah. But see Rich since the boom is out there and most of the companies are already you know have hopped onto the bandwagon of AI. Are there any real world success stories that you know you were inspired by? Real world success stories. You know I write about a lot of them on my blog at tyranni.com. I’m constantly blown away by success stories. I mean I don’t want to point out any one in particular. I mean, I would go there. I’d point people there. But companies are doing amazing things right now with agents. The agent opportunity is the biggest and it’s where you basically are having agents do the work of actual humans and you’re able to scale. We had one of our exhibitors at past IT Expo show. They were telling me that they had an agent that they deployed in an airport. It was a trial really. airline had a strike or the airport had a strike and the airline was was doing the deployment and they they were deploying a small amount of agents just to see how it worked. But then all of a sudden they were flooded with thousands of phone calls simultaneously and they didn’t have the agents to take the calls, human agents. So they had no choice but to use and the AI agents handled the job well and the airline realized that this was these AI agents really important because you know scaling is a big problem and again contact centers are another area where scaling is a big problem and we’re going to see agents there and so think about you know if you have a a contact center where the number of calls around the holidays spike 10 times the normal amount can you hire 10 times the the normal amount of agents? Yes, you can. you can outsource, but using agents is another way of doing that in case you have spikes throughout the year. So that that’s I guess one example I can give you, but there are so many others. I remember yesterday I was marveling at just how many I I was covering. So yeah, I’d point you to torn.com to keep tabs on that. But then, you know, that’s the positive side of it, you know, and nothing comes without the flaws. And you probably must have witnessed, you know, technologies evolving but with their flaws of the AI as well. So in your experience, where have AI agents fallen short, you know, in perhaps overstepping their bounds or causing unintended consequences? We’ve seen the hallucination issue. So that’s the unintended consequences, but I’d say we’re not at 100% par with human to agent or agent to human yet. We’re we’re getting really close. You know, there are some technologies that are maybe 99 plus%. But trying to approximate uh either humanlike quality or getting to a point where customers or or people on the other side are comfortable working with the agents. I guess that’s got to be the next frontier for at least for the ones that are are humanfacing. But in terms of agents that are working within organizations, there are still challenges like I’ve used just give you an example of uh something I’ve done with um open AAI their chatbt operator agents. I’ve designed agents to populate databases. You know, aggregate data from one database and then put it into another database. Interestingly, I had company names, but I didn’t necessarily have the username. So I said, you know, find these find usernames for these companies meeting my criteria and put them in a third database, a CRM database with the proper coding. And uh, you know, to me, I love sometimes to be on a call like this or just a a meeting and and just out of the corner of my eye, I can look at one of my monitors and watch these agents populating databases. And so I was looking at it and I I saw it added a name, John Doe, and I was like, that looks really artificial to me. But I was like, okay, whatever. And then then I looked at it again, it said Jane Smith. And I was like, “This these are these names are just don’t look real to me.” So I I paused it and I waited to the end of my call and then I went to it and I said, “Okay.” I looked and there was no John Doe at the company. There was no Jane Smith. And and I said, “What what is this John Doe and Jane Smith? Where did you find these names? They’re not at the company.” And it came back to me and it said, “Based upon your request, I used hypothetical data.” I was like, “Oh my god, I never told you to use hypothetical data when putting data in my CRM. I didn’t want you to do that.” So now my prompts include do not use hypothetical data if I’m making an AI agent. You know something that never would have occurred to me and it probably wouldn’t have if you were asking a human to do that job. It wouldn’t have the human wouldn’t have done that. But you need to change the way you give instructions because the AI really wants to please you and it goes too far sometimes. In this case it went too far because there were no names that matched the criteria that I had and I didn’t specify what to do in that circumstance. So uh you know a bit of a human side towards this entire AI boom thing rich as more organizations adopt AI agents to manage cyber security and network management where do you see the line between letting AI handle autonomous decisions and ensuring that humans remain in the control? So I gave an example of just kind of human oversight and and that’s kind of where we are in these early days. I say, you know, as we’re we’re allowing these agents to run wild and do more and more within our organizations, we we need to keep an eye on them and and just make sure they don’t actually go too far and they don’t run too wild. Over time, we’re going to see multiple AI agents from different companies able to police the work of one another and that’s going to make life easier on the human. And then the humans will be flagging something that looks anomalous or looks weird. Like for example, let’s say you run the output of one AI agent through two others and one of the other two agents thinks there’s a problem. Then maybe the human will flag it at that point. I think that’s where things will ultimately go, but right now you’re going to get the output from one agent and you’ve just got to keep an eye on it. Maybe the end of the day you’ll look at the logs of all of the interactions and you can run that through another agent and say, “Okay, what what looks weird here? What what is there what problems could there be?” So, we’re we’re in early days and you do need to keep an eye on things to make sure that you don’t have excess liability because you’re using agents that don’t have the human I guess the guard rails that that humans naturally have in the workplace and u you know it is it is a hot debate these days that how you know every industry eventually is going to be replaced by the AI and you know everyone is concerned with the recent Meta update on how you know Meta’s AI agents are going to take over the ad agencies and that is you know very concerning for most of the businesses out there quite irrelevant over here but then what it tells us is that eventually every manual work is going to be automated one way or another every disruptive technology that we’ve dealt with has eliminated a lot of jobs but it’s also created a lot of jobs and I like to use Microsoft Excel as an example uh Microsoft Excel does the work every day. It does a work of probably millions of accountants and yet the number of accountants continues to grow and there’s an accounting shortage in the world. And that’s just an example of because Excel has made it easy for anyone to be an accountant or or for a lot of accounting tasks to be given to people that maybe were not at that level accounting level. They made it so much easier. It’s now easier to start companies. It’s easier for one accountant to do work in multiple companies. So the cost has gone down. So now starting companies has become less expensive. And so because starting companies is becoming less expensive, we have more companies and because we have more companies, we need more accounting work. And so that’s traditionally what’s happened. Does AI present a new opportunity and challenge? Is it scary to me? Yeah, it’s a little scary because like you said, AI can do so much and it can replace jobs and there is already talk of one person billion dollar valuation companies where you don’t hire anyone. You do everything with AI and you don’t hire a single person. And certainly the meta example that you give me is is or that you just gave is really scary because you know as Mark Zuckerberg has said they are moving to a solution where you’ll just tell it what you want what kind of person you want to respond to advertising and then you just don’t do anything and it just goes and find makes its own ads experiments. It determines which ad is the most effective for your audience. It determines where to run that ad, what time, what location, etc., and then you just get the answer. So, yeah, that can wipe out ad agencies, creative agencies, marketing agencies, right? They’ve all now potentially been wiped out as this technology from from Meta gets more widespread. So, what do those people do now? And I don’t have answers as to what those people are going to do. I mean, they they’re going to have to make the decision. And I’ve been saying I’ve been hearing that, you know, you’re not going to get replaced by AI. You’re going to get replaced by someone who knows AI better than you. And and we have to hope that that’s the case. And to the extent that everyone can learn how to use AI and adapt and and become more proficient, we’re we’re actually we have an AI agent event we’ve launched at the end of September the September 29th through 30th um 2025 in Washington DC. And we have a certification there and it’s certification that you you know to learn the basics of AI agents and we’re doing that because we want to help people to keep jobs to show that they are trying to stay relevant in a market a job market that probably has not seen this much change since you know blacksmiths were eliminated when the when the automobile came out and you didn’t need horseshoes anymore and there were no more blacksmiths right like some massive change like that I think everyone needs to it’s called upskilling everyone needs to upskill and hopefully, you know, there’ll be lots of new opportunities for people. There will be for sure, but I don’t know how many and if it’s enough. And that’s what we have to hope for as a society. But I mean, the the flip side to that, and I’m sorry I could go on just on this topic for a while. The flip side to that is that theoretically it’s going to be a lot less expensive to buy things because automated farming, automated food production, automated clothing manufacturing, right? Theoretically, this is going to bring the price of everything down. So now we’re going to have access to more because things are going to become lower and lower cost. That’s the theory. So we won’t need as many resources to acquire those things. So it the world is going through multiple transformations and where this ends up is is unknown but upskilling historically has been the way to combat these disruptions. Absolutely. Absolutely. And if rich we look in the mergers and acquisitions industry and their readiness towards the AI, do you think that AI is now a hot commodity there as well? AI is helping companies determine which other companies to acquire for sure and it’s also disrupting just in terms of industries the types of companies that are coming into those uh in terms of AI first organizations that are getting funded. I’m going to say that AI is playing a part every part of business from I say from banking analysis to putting together documentation to putting together models uh business models of the of the target that you want to acquire uh AI can put together business models that humans could probably do but they do it a lot faster and they can take into account a lot of factors that humans just wouldn’t have time to do. For example, if you’re looking to acquire a company and you have a business model and you believe that in the next 5 years that you know the company’s going to produce this much revenue every year or the earnings are going to be so and so every single year, you know, you you typically would have that model and and that’s what you would base your value on, your valuation of buying that company. But now, you know, there are all sorts of factors that could play into the profitability or the earnings of that company. And you know it could be geopolitical factors, climate change factors, you know, wars, tariffs, embargos, etc., etc., right? All these other factors. Now, what you could do is have AI model out the likelihood of these scenarios. And then how each of these scenarios in aggregate can affect the earnings of this company over 5 years. And so now you’re able to have a much finer tuned set of earnings numbers for the next five years that you you couldn’t easily do beforehand. and you use those numbers to make a better, more informed decision. So, I mean, just like AI is being used by students, right? You know, 10-year-old students to help them learn, to help them make college papers as they get a little bit older, you know, it’s it’s finding its way in everything. Yeah. No, that’s true. That’s that’s really true these days because, you know, you’re just finding AI at all of the crossroads and there’s no way that, you know, you can just avoid it at this point in time. But then you know uh rich given your exposure and experience with both the industries of be it banking finance or the tech side of it how do you you know keep the balance or how do you feel that your insights are different in comparison to someone who’s just focused you know in or towards just one industry. Well I mean I I guess there’s two sides to it. I don’t I don’t have the ability to go as deep on the things that I’m working on. And right, that’s why I always try to I try to get up earlier and stay up later so that I’m I’m as immersed as I can be in all the different things that I do. But I I think my unique insight is I’m very fortunate in that I’ve done a lot of uh writing and communicating over the decades and that that communications is quite often taking very challenging information from a founder who doesn’t necessarily know how to describe their company very well and then put that into something that’s interesting enough for someone to read and something that they can absorb very quickly. And and it turns out that when you’re when I’m helping companies raise money or I’m helping them sell their company, that’s a pretty important skill to have to not only understand the technology pretty quickly, which you know, because I’m I’ve kind of grown up in it. I understand, you know, most technologies pretty quickly. I can get up to speed, you know, very fast. I’d say faster than most just because I’ve lived it. and then on the other side explaining it out to potential buyers or or to VCs or to family offices or to whoever’s trying to make that buying decision, explaining to them why it might might be a solid investment or or what they should be looking for. So, I think that gives me kind of a unique perspective um between those those two hats that I wear. And you know, it’s exciting to be at the crossroads of both areas because again, disruption uh banking is getting disrupted by AI, uh media, technology, right? Everything’s getting disrupted and it’s just fun to watch the intersection all at once. Yeah. But then you know also how does it feel like to be at the core you know of providing a platform like it expo you know tech super show and uh connecting with people from all around the world. I grew up in the trade show business. My dad at this same company TMC started a trade show in 1986 the first call center show actually in Atlanta Georgia in 1986. And you know, I I grew up building these live markets where people would come to buy and sell where, you know, we’d have exhibitors that are trying to sell things and then we’d have buyers looking to buy things. And to me, it only gets more exciting to see the change happen in the market and how it’s all evolving to be different kinds of solution providers that the buyers are looking for and and you know watching that evolution you know maybe 25 30 years ago when voiceover IP was much newer a lot of people were not sure if voiceover IP was right a lot of companies weren’t sure and we saw you know it took a quarter century for them to understand that voiceover IP and these communications networks running over their traditional data networks or their IP networks made sense and now we’re evolving that to have AI and and on top of that we’re going to have agents and generative AI and and so just watching the evolution of the types of technologies that companies are looking to buy has been really fun and exciting like in 1986 keep in mind this was a few years I’ll go back a little Earlier in 1982 when our company launched the first publication in the call center space, first magazine, the traditional way of of keeping your contact center or keeping your customer records was on a piece of index card inside of a metal box. That was 82, 83, 84. Then the PCs came along and we had mini computers and all that and we eventually evolved the technology. So like I I got to see from index cards where humans were on index cards and then it went into you know went into a maybe it was a mainframe and then from the mainframe it went into a mini computer and then the PCs were in the ‘9s and all of that change and watching the different exhibitors and the different software companies has been great and then you know we had this whole big run of cloud companies and now we are seeing the AI companies and are they going to replace the cloud companies are they going to be the same or a A lot of that we don’t know yet, but it’s really exciting. So, to me, bringing buyers and sellers together, you know, hundreds of exhibitors, thousands of people from around the world looking to buy things. Nothing is more exciting to me than that. But the foundation to all of it was probably set by your dad. Yeah. In 1972, uh, the company was founded because he was a chemist. And at the time, there was an oil embargo because there was a war in the Middle East. Does that sound surprising to anyone? war in the Middle East. Um I’m kind of making a joke because yesterday that the Iran was bombed by Israel. So it’s just, you know, something never changed. So there was an energy embargo. There was an oil embargo on the United States by the Arab countries. So all of a sudden the US had that we went from having cheap oil to no oil or limited oil. And so the way we had to work changed. So my father realized at the time that the way we manufactured was going to evolve. And at the time if you painted a car for example, you put it through an oven and the oven was open on both sides. You heated up this car to dry the paint and all this energy was wasted and that we couldn’t do that anymore. So that was the birth of new pigments being put inside of paints where instead of having to use heat to dry them or to cure them as it’s called, you’d hit them with ultraviolet light. And now if you go to a nail salon, it’s very common now that you’ll see ultraviolet lights and nail salons. That technology was the first climate friendly or green technology invented in the 70s, early ‘7s. and my father launched a whole bunch of magazines to teach companies about these new coatings and these new paints. That actually was the birth of TMC back in 1972. So that’s where he started. And you know, for me, I I was fortunate enough he allowed me to grow up in the company. He allowed me to, you know, I was putting stamps on envelopes when I was six. I started getting paid when I was nine. So I learned the value of work and I learned the value of making money. And I, you know, thank God I had that opportunity. And so I I owe him a debt of gratitude. Yeah. and not becoming a chemist after your father. I never was a fan of chemistry. To me, it was just a boring it was boring to me for whatever reason. It was just a lot of, you know, especially the chemistry classes I had at engineering school. This it was more about numbers and and conversions than actual what chemistry could do. Computers to me fascinated me because when I was working for my dad in the mail room, my parents got me a computer as a present and I was using the computer and I learned how to program and then we had hired a consultant and they put a computer in our office but they never finished the job. So I had a computer there using the same language basic which I had learned at home and they just didn’t do any any of the programming. I was like and I said to myself, well wait a second, I know how to program maybe I can do this. So, I just figured out how to program and I automated all of the work that I did in the mail room and I was able to put all the names inside of the computer. And then from there, we were able to outsource the mail room work to the printer because they had an automated way to put the labels on the magazines. And so, I automated myself out of a job, but I then became the IT guy and I and I I didn’t have to lug around magazines anymore. It’s a much better job. But I said to myself, “Wow, look what computers can do. Look at just how automation for me allowed me to change my life.” and and it just blew me away. And at that point, I wanted to be an engineer. I actually I just wanted to learn how to program computers, but at the University of Connecticut at the time, you had to be an engineer. So, I did a lot of a lot more of the math conversion stuff that I just told you about that I that I really didn’t want to do, but I did all that. Uh because they kind of forced me to. But yeah, my love of technology just came from that. Yeah. So, now we’ll talk a bit about the future of the AI agents that, you know, we supposedly all of us think about. So if you look at the next 5 to 10 years, Rich, do you believe that AI agents will take over most of the network and cyber security decisions and would be able to operate with, you know, little to no human oversight at all? In most cases, I’d say we we’ll get to in 10 years, we’ll be at 99% plus without human intervention. And the reason for that is that you’re going to have different AI models that are different enough that they will act as the human oversight. And because the cost of AI is plummeting, cost of the hardware is plummeting. As long as the electricity is there, right? That’s our limiting factor. Well, we have the electricity. As long as the electricity is there to power all of this, then the AI becomes near becomes closer and closer to free. And so as that happens, then multiple models attacking the same problem from different angles and maybe comparing their results. Ultimately, there could be a human oversight when there are some conflicts or there’s some issues which will get filtered up. there’ll probably be some kind of mechanism or some kind of alarm to to send to a human in certain cases but for the vast majority of situations especially in I mean this is for all of a company but cyber security especially because you’re dealing with something that is so time critical and timesensitive because you’ve got all these tools that the good guys are using the bad guys are using too so while they’re attacking you you’ve got to respond as fast as they’re attacking you and if you don’t you can’t say okay we’re going escalate this to a human. Oh, the human’s sleeping. Well, in the morning when he or she comes in, they’ll deal with it. I mean, that that’s not a solution. Or you could just have the human look at it and then analyze it if they’re 24/7. But even then, you know, it doesn’t take much time as these computers get faster and faster to do an enormous amount of damage critical systems. So, you’ve got to, you know, you’ve really got to um fight them as fast as they come in. Have you ever played the game Fruit Ninja? Yeah. You ever you’ve played Fruit Ninja? I think of it in my head as fruit ninja, right? Fruit’s just I haven’t played this game in years. I don’t even know why I thought of it. But, you know, fruit is just coming at you and if you’re not slicing the fruit as it comes, you’re going to get hit in the face with a mango or a watermelon. And you don’t want to be hit in the head with a watermelon. You want to have your sword and you want to hit hit that watermelon and hopefully it goes on both sides before it hits you in the face. And that that’s for and it’s not just that. It’s for everything. I mean, I’m thinking of these AI agents as like fruit ninjas and they’re just like problems are coming in and they’re just chopping the problems and sometimes you’ll have multiple ninjas, right? Like I said, right, to to solve these problems. Yeah. No, that’s true. But Rich, how is banking and finance going to be affected with the emergence of AI and what has been happening, you know, over there in that industry? What what is your insight in that department? It’s changing everything. Right. So, the area of banking which I’m really not involved in is investment management. So that’s an area where I’m seeing I’m seeing a lot of disruption just because AI can pick stocks and can determine what are great investments often times better than a human. But then you’re getting into the modeling of numbers which you know a lot of these a lot of the work in banking is modeling of future revenue and then also and and earnings and also trying to present data and AI is really good at all of that. And I I think you you mentioned the CEO of Goldman being a DJ. If I remember correctly, the CEO of Goldman also said that uh when they put together a document for a company going public which is called an S1. I think it was a tweet he put out if I’m not mistaken. I remember reading I think I was in the gym when I got it. I I was kind of blown away. I was in between sets looking at it. I was like wow. I think he said 90% of an S1 can now be put together by AI but still 10% needs to be done by a human. Obviously that 10% is really important because you can’t make mistakes. But 90% now can be put together by AI. And so at least at this stage we’ve gone from having to do 100% of the work to 10% of the work. Crucial 10% of the work. And you need a lot of experience to know how to do that 10%. And the problem actually is this is this is another problem and that is that a lot of those entry- level jobs are kind of drying up because AI is doing those entry- level jobs. So the question becomes how do you gain that experience? Well, that that’s a problem for another podcast I guess. But the point is that so like you had said earlier, I think you said about 20 30 minutes ago that you know all the manual work is going to get done by AI and so whatever is considered manual and so what is considered manual? Well, putting together documents is manual really it is. We think of it as high value work because it was but is it is it high value work if the computer can do it? And the answer there is that more and more of this work is just becoming what we will consider manual labor where the computer will do it. And you know the one thing that didn’t get mentioned is okay the human needs to do that last 10%. Well what happens when you attack that same problem with three AIs or four AIs or five AIs or six AIs and now it’s 99%. Is it 99%. I mean is that where it is? Is it 100? I those are unknown areas and and that’s kind of where things are going with everything. Again, it’s not just bank, it’s every job, right? Legal, it’s the same thing with legal. As good as a lawyer as you can hire, they cannot store the the entirety of every legal case in their head. Give you information that is as well-rounded as the proper al and I’m not saying the algorithms are there as the proper AI algorithm can do because it doesn’t have access. he doesn’t have access to that information and maybe given enough time he or she could but they it it’s just the computer has instant access to it and so for for all these jobs you know medical diagnosis we’re already seeing that um AI can diagnose cancers better than the majority of doctors so that’s another area radiology uh and it just goes on and on from from there banking is getting disrupted all of banking is getting disrupted M&A is getting disrupted we all are getting disrupted. So, it’s unknown where this is going in terms of where those jobs are going to be and what those people are going to do or how we’re going to evolve, but um upskilling seems to be the best thing you can do right now. And whether we’re all going to have our own startup and that’s the future, I I don’t know. But it’s going to certainly be an exciting time. Whether that’s a good exciting or a bad exciting, I don’t know. You know, depends on your role in the organization, right? Yeah. But also you know one surprising thing about about you is that you have an um you know academic background in software engineering but then you know from the past five years you’ve been an investment banker as well. So how did you come across this particular industry? I had no interest in doing this at all. I this is not something I if you had asked me years ago I just thought of it as another area that I had nothing to do with and I never will. But my clients asked me for help with with mergers and they kept saying can you help me with this? Can you help me? I want to sell my company to this group. I want to sell to that group. Do you do you know people? Do you know someone here? Who would you sell to? What would you do? How would you plan for the future? Help me with my road map. I want to make my company valuable so I can sell it. And so I I just was doing that kind of work and then the next thing I knew they wanted me to do the whole job of selling the company for them. So I said, “Okay, I’ll do that.” And then someone asked me, “Can you help me raise money?” And then when I started doing the money raising, I didn’t even know I needed to be a banker. I learned I had to become an investment banker. So, I had to take a whole bunch of courses and and do a lot of studying and learning and you know, I did all the courses and and and so that’s kind of how I got here. I didn’t I didn’t plan to be here. I didn’t want to be here. I I love it now. I love seeing the other side of technology because to be honest without the rich companies, without the rich people to give their money to venture capitalists to put into tech companies, we would not have any of this innovation. All of it, all of it, all of it. Google is an example. I mean, the computers, all of it, all of it came from venture capitalists making decisions to spend rich people’s money to try to get a great return on their investment, betting on something that a lot of times seems crazy. Like, I’ll just give an example, Uber. I mean, 20 years ago, everyone in the if you ask anyone in the world, should you get in a stranger’s car? They would have said no. What are you? Obviously not. You learn that at age four. don’t get in a stranger’s car. And and someone decided to build a business based upon people getting into other strange people’s cars. And it became, you know, tens of billions of dollars, wiping out the taxi industry. Now it’s becoming a massive part of the food delivery industry, grocery delivery, all sorts of deliveries because someone thought that they could do something differently. and in a way that we were told we couldn’t. And VCs were there to put money into these companies. And they also had to go against what they were brought up to believe. And thank God we have these VCs. And getting to work with them, I got to say, it’s it’s some of the most exciting work I do. Some of these VCs are the nicest people on the planet. Others, you know, they they’re like any other industry. Others can be challenging and and difficult and and, you know, challenging to deal with. They maybe they had a bad day. that a bad investment whatever it is so you know depends when you catch someone but as an industry thank god for these people and you know when we look at why is America such a land of opportunity and it’s not the only land right you know India and other countries are doing pretty well but thank god we have this very strong base of investors that have helped propel constant innovation yeah and hopefully that continues you know every day is a new day and as long as you keep learning one can you know keep evolving I believe moving On to the next question. Rich, if you were advising a CEO or a CIO today about integrating AI agents into their organization, what’s the one hard truth they need to face when it comes to AI autonomy? So, get okay, AI autonomy is the eventual goal. I I like to call it the autonomous enterprise. Start slowly, start with the agents, get up to speed, learn, learn, learn. Like you said, improvement, improvement, improvement. Every company has to focus on becoming AI first. You have to. If you’re not, you’ll be disrupted by someone who is. Start deploying the agents. Test, measure, manage, improve, iterate, test, measure, manage, improve, iterate over and over and over. I mean, we’re at a crossroads. I actually wrote today about, you know, the next challenge is going to be getting these agents to work with one another. That’s that’s another area that eventually we’ll have to figure out. They say that it is going to be the next human resources, the next HR, which which makes sense because you’re going to have these different agents as humans. Right now, they’re doing the work. So, how do they work with each other? Do they what if one makes a mistake? Does it get amplified through the other agents? So you can’t make a mistake if it’s a critical agent. How do you double check that it’s not making a mistake because now it’s going to ripple through your organization. So test, measure, manage, you know, just stay on top of it and understand that we’re going through a massive massive shift. And if you if you’re not ready for this shift, you you may get wiped out. May maybe now is the time to sell your company to someone who can deal with this. Figure it out. Hire people. Um there’s going to be a a big industry of companies helping and something we’re looking at at Apex also is you know from IT services how do we help with AI um to evolve to help these companies evolve because not every company has the stomach to do this even if they have the money that you know it’s it’s managing and code and it’s liability if the agent makes mistakes. I mean there’s a lot here. How do you deal with all of this? It’s a whole different business model but you have to do it. You have to do it. So start slowly but then ramp up fast. Yeah. Yeah. But you know um given that we are um somewhat way ahead in it, do you do you believe or think that there is any misconceptions out there that are already known and that is something which can be avoided by sea at this point? I think the biggest misconception is probably that AI is not going to affect my business. Um Oh yes. Yeah. Unless you’re and there maybe if you’re um if you’re doing plumbing work or there are probably some areas where where you’ll be okay you know even truck drivers that you know 10 years ago they would have said oh I’m not going to worry I’m always going to have a job well there are autonomous trucks I mean they’re not widespread there are issues there safety issues etc that will get worked out but you know thinking that you’re in a role you don’t have to worry about it when I when chat GBT first came out a few years ago when it no was I think it was version 300 when it became more popular and I was telling my doctor friends about how you can do a diagnosis with it. They said, “Well, I don’t want to work. I don’t want to deal with that. I’m going to hopefully I’ll be retired before this affects me.” That’s my social circle. A lot of them were saying things like that and and I just think that, you know, you need to realize this stuff is getting fast or it’s getting good fast and it’s going to disrupt what you do. Your company is going to get disrupted. Uh maybe not in the next four months, but within 18 months to a few years thereafter, things are changing. You need to stay on the leading edge of it. You need to be educated and you need to devote time every day to staying on top of the different AI tools, the different things that are happening. You have to be on on X. You have to see what’s happening on LinkedIn. Stay in the news. Whether you’re following my blog, whether you’re following, you know, there are a whole bunch of sites out there which are really good. Techrunch is good. CNET’s good. ZDNet is good. There’s a bunch of great sites out there where you can follow along. Venturebe can be good. It’s more on the investing side, but even general business publications, you know, Wall Street Journal, Bloomberg, you know, you got to stay on top of this stuff. And if you’re not doing it, you will not make it through this transition, you know, in your exposure um given that you work with um different sorts of people, be it in tech or in investment banking. Have there been moments when you had to deal with those traditional mindsets, you know, who are still keeping pace up with their legacy systems and still reluctant to move on? Is is there still a lot of room to be improved there with those people? There are we’re still using floppy discs to maintain the the airplanes in the US, right? Our air traffic control is floppy disc based. And so the plain truth is that you know the programming language Cobalt still runs a lot of a lot of technology in the US and in the world. I think I wrote about it a few days ago. One of the investment banks has made a tool that converts cobalt to newer languages. Thank god that’s good. You know, co AI is helping evolve the old systems into the new. But yes, and that’s a common problem and it, you know, if if a company is, let’s say it’s 20 years old and it’s got systems that have been built over 20 years. I feel for them because converting that and and I’ve seen it like for example, my IT consulting company, I’ve seen some of the clients there, they work certain ways with certain software in that you wouldn’t think that that software should do what it does. And it they’ve kind of tweaked it and they’ve kind of made it where their um their CRM is based on on something that shouldn’t be a CRM. and it works for them, but how does that become an AI first company if their tools are not as flexible enough to work with the AI? And and so this is painful. This is a painful process for companies. And that’s why when you talk about M&A, I mean, it’s not just the culture, what are the tools that are being used right now. Um, you know, I think Salesforce has done a good job of evolving their systems to have AI on the front end. So, if you’re based on Salesforce, for example, on your CRM or some of these others, you’re probably in in better shape than if you’re using some other systems um that are not as upto-date with AI. And so, yes, the problem is these legacy systems are not going to get upgraded right away. And this is why someone like Mark Andre is betting so much on disruption because it’s a heck of a lot easier to give money to someone to restart or to start a company that’s AI first than it is for someone who has a 20-year-old system that their company is based on and to figure out how to turn that into AI first. It’s not easy. And so, like I said, this is big big disruption for companies. It’s going to be scary for some people and and rightly so. Are there any people in your close circle that you’ve given this advice to? I told my doctor friends, get up to speed on it, right? I I took out chat GPT at dinner and I and I I said, give me some of the, you know, tell me about the symptoms of some of the patients that you’ve seen this week and I put the symptoms in chat GPT to get a diagnosis from Chat GBT and I showed them and and they couldn’t believe it. This is again when it first came out it just you know right after they had immobilized they couldn’t believe it and you know you saw the light bulb go off above their head and and they got it and so I mean those are that’s a good example but you know I have other doctor friends I don’t know why all my friends are doctors I have a lot of doctor friends apparently now that I think about it another doctor friend that I you know I was talking to her about cyber security and she said that’s why I keep all my f my patient files on paper and folders so I obviously said what if there’s a fire and she said We keep them in fireproof cabinets. And I said, “Well, what if there’s a flood?” You know, a pipe burst. It happens. What What are you going to do? Do you take a picture of them in like I I didn’t go down there. But anyway, I think my social circle, if I’m going by my social circle, I maybe this maybe they’ve aged out of AI, I guess, maybe. But, um, when I live in the tech world, it’s a totally different thing, right? So, when I’m when I’m with my tech colleagues, it doesn’t matter how old you are, they all know, right? Unless they’re retired. Yeah. Yeah, I don’t care if you’re 65 and in tech or 70 and in tech, you know, you know that you’ve got a AI is the thing and you better learn it or or you’re you know, you’re just not going to be around. You’re going to assuming you’re still working at that age, right? If we look at the job numbers in the US, as an example, our unemployment rate has been pretty low for a while. So obviously a lot of work that was done by humans is being done by AI, a ton of it, a ton of globally. Yet we’re not seeing mass unemployment. We still do have obviously jobs that are being created despite the fact that you know AI I’ll throw a just a round number out there. AI is probably doing the work of 20 million people right now, right? 20 30 40 I I don’t know. It’s got to be at least 20 million people, right? Yet we’re not seeing an unemployment crisis in the world. So and we’re not even seeing a huge uptick. So point there is that not everything gets replaced overnight. things are going to evolve. Even if I was to reinvent General Electric as an AI first company and have a much much better solution than General Electric because it’s AI first, General Electric still has customer contracts, five-year contracts, 10-year contracts, they have customer relationships, they have a reputation, they have so the techn we have to be careful as a as an industry to not get too far ahead of ourselves. I think it was Bill Gates who said that, you know, we overestimate every technology in three years, underestimated maybe at seven if that’s 7 to 10. So we may be overestimating, but then what’s going to happen is this this stuff’s just going to happen when we’re not talking about it as much and maybe that’s going to be 3 years, 5 years from now, whatever that is. So that means what does that mean? That means we’re in the opportunity phase. That means that now is the time as an AI first company for you to start a company that’s AI first to try to disrupt whatever industry you think you can disrupt. We’re seeing a lot of startups and that that to me is the most exciting part of my career is that I I get to write about these cool startups and you know whether it’s healthcare or whether it’s um finance or legal um it’s great. It’s great. And and Rich, you’ve been, you know, writing for quite a while now. And um I was just curious what still keeps you passionate about writing given the advent of, you know, Chad GBT AI and lots of other writing tools. Do you happen to have a digital twin of yours who is doing it now? I love communicating, but I don’t love portions of writing. There are certain portions of writing that I despise. I’m not a huge fan of making introductions to things. I’m not a huge fan of making conclusions to things unless I’m really passionate about that topic. AI gives me this unbelievable ability to take raw thoughts and raw notes and raw ideas. And sometimes I can I can put a fine point. I I wrote something this morning where I said, “Okay, I like this research report. I’d like to start out an article like this, the following.” And I gave it the first few paragraphs and then I said, “Okay, and here’s a research report. Keep going.” So something like that. I’ve never had that ability before because I would have to go through the whole research report. I would have to determine what the best part is. Then I’d have to second guess myself. I don’t want it to be inaccurate. So, is this really the best part? Is it not the best part? I think it’s the best part. Right? And that that could take me five hours just going through the research report. But now I have a point I want to make and I have the research report. I’m able to disseminate that information in a way I could never do quickly before. And because I couldn’t do it quickly, I might not do it. That’s a huge change where I’m able to share information over the last, you know, decade, the decades of information I have and the decades of viewpoints in a more succinct way, at least time-wise for me, but with this the full breadth of information to support predetermined notions of things, if that makes sense. So now I’m able to do this work that I couldn’t do quickly before, but now I can do it more quickly and I can get the information out there and I can start the conversation with my community and move on. And so that to me is great. And that is worth getting up for because I’ve taken the drudgery out of my own life. The parts of my life that I despise. Like, you know, like I said, certain parts of writing I just hate. All the parts I love. I can do the parts I love and I can allow the AI to do the parts I hate. And I proofread it and I edit it, right? And I, you know, sometimes the proof reading is better than others. There are occasional hallucinations that I may or may not pick up. I do my best to get them all. But again, you’re dealing with AI. It’s not perfection. But, you know, if you can get it to 99%. And you know, sometimes on my AI, I’ll have AI help me with something and then I’ll run it through if it if it’s something where I think it’s really factual issues that could be a problem. I sometimes have a sense I’ll run it through a second AI as a proofreader. In the beginning, I didn’t have any clue about writing. Even though I took classes where they taught us how to write, I was too much of a know-it-all in college because I said, “When am I ever going to need to write? Why would I ever need I didn’t really even pay attention the way I should have in in college and in high school to my writing professors who in hindsight were really really good but I didn’t pay attention to them at the time so I didn’t have the experience but then I was kind of on the job training and over the years I I kind of understood I think you know if you especially nowadays people are so pressed for time I try to put takeaways at the beginning of all the things I write and I think the takeaways are important because I know even myself I can’t read everything I want to read but at least if there’s an interesting headline. If I can get five takeaways out of it, I get a sense of what the article is. I’ll just move on. If I if I want to read it, I can. Sometimes I’ll save it for the weekend. Unfortunately, I think that’s where the world is going. We just don’t have the time to read as much long form content. Yeah, that that is what I was coming to next, right? That people don’t have that kind of attention span anymore. They would rather, you know, watch a 30 seconds reel over reading something. So, you know, what then keeps a writer motivated to still be able to keep doing it? These are really interesting questions. I don’t know how to, you know, this is this question can be really applied to a lot of work. AI is changing it. I happen to hate certain parts of what I was doing. I’m lucky. I I’m lucky because AI can do those things for me. That’s what keeps me motivated. And I know some people are going to read. Not not everyone, right? But not everyone read anything that, you know, the best read articles still. I’m sure not everyone has read Harry Potter, right? It’s very, right? Not everyone’s not everyone’s read everything. So, you know, not everyone’s going to read it. So, you do it because you want it to be complete, but you have to kind of provide it. I mean, it’s almost like, you know, after I’m done writing, I’ll tweet it. I’ll put it on LinkedIn. So, not everyone on Twitter is going to want to come to my blog either, right? So, they’ll stay on Twitter or they’ll stay on LinkedIn. So, the question is trying to get the information to them and in the area or the form that they they prefer to have it, I guess. Absolutely. Well, uh, Rich, that sort of brings me towards the end of our conversation and the last question. Rich, over the years, what’s this one, you know, hard learned lesson that you’ve had in your life, be it, you know, back in your academia or your professional life? I love it when my guests have to think on that. Well, I want to give you the most accurate answer. Um, I I think understanding disruption and when your business is going to dis get disrupted and what it’s going to do to you. I mean, just for my own business, we used to have a thriving magazine business that went to a thriving online business and now that’s turned into more of an event business. And so seeing disruption affect your business and not thinking that you are smarter than the global economy and trends in the market. And and this is just from from my own running companies. You just don’t you don’t know what’s coming around the next corner. And if you don’t think you’re up to the task of dealing with the disruption that’s coming in AI, this kind of comes full circle. If you don’t think you’re up to the task of dealing with how it’s going to affect you, you need to make a decision soon. either you’re going to sell your organization or you’re going to make a conscious decision that you’re AI first because nothing I mean again except for the plumbers and some of those roles you’re not going to you won’t be around that’s a hard truth that that I learned some time back and you know you continue to learn it you continue to learn it no every industry gets disrupted the number of companies that get wiped out gets increased every year more companies get wiped out the percentage of companies existing companies they just don’t make it through and the pace of innov innovation continues to increase. So it becomes tougher and tougher for a company to last through the ages. So and so you need to either you would adapt or die, you evolve or die. And and that’s just the name of the game. Yeah, absolutely. Well, Rich, thank you so much for joining us today and for sharing your valuable insights for our listeners and viewers. Thank you. I appreciate it. [Music]