Harnessing AI for Maximum Performance with AI Expert Kevin Surace of Appvance.ai

Out of the sky one day. We've been working that for 70 or 80 years, right. And I started in the AI space inventing the first virtual assistant. Her name was Mary. And eventually all of that got licensed for Siri and Cortana and Alexa and all of the other virtual assistants. You can look at the advent of large language models and say, what kind of company can I create leveraging that technology? How can I use it to come up with something that really solves someone's pain and has enough of a moat around it where you aren't going to have a hundred other companies do it?

Welcome back to another episode. Today we have a true entrepreneur's entrepreneur on the show. We're going to go into some really cool stuff and really talking about the value of building over your life, the complete and thorough journey. Today we have Kevin Surreys. Kevin, thank you so much for coming on the show.

Yeah. So happy to be here, A.J. thanks for having me.

I am very excited. I mean, I've been just like going through this like one cool tech company from my past to another to now really cool AI. You're the CTO and Chairman X CEO of Advance. You're a well known public speaker on the board of tons of companies like Token and many, many more. I mean, when people talk about a successful entrepreneurial journey, you're kind of that picture and you have the Inc. 500 or the Inc Magazine to back you up on that. Can you maybe talk a little bit about like where you are on your own entrepreneurial journey these days?

You know, there's so many ways to define the entrepreneurial journey as, you know, and as we had talked about an earlier point, some people really talk about it as if work is their life. And that's it. That's all there is. There's no question that for many entrepreneurs they make work their entire life and it becomes who they are. But you know, I gave a speech on AI, I don't know, 10 years ago or so at a TEDx TEDx Orange county, and I talked about a little bit of the dangers that over millennia we have sometimes become known as just our work. And I go all the way back to Jesus time. We really don't know his father, his earthly father's name per se. We call him Joseph the Carpenter. I don't know what his last name is. He's just the carpenter. So good grief. You'd think you'd have a last name. Maybe you'd have some other things he did in life. He's just Joseph the Carpenter, it's AJ the podcaster or whatever. Right. So we should be known by, for much more than that. And that gets into a whole cycle of what brings you joy. And clearly there's joy in success, but there has to be joy in failure. If there's not joy in failure, don't be an entrepreneur. That is a ridiculous thing. All the time they say, well, you know, am I born an entrepreneur or this or that? I said, well, entrepreneurship isn't for everyone, but if it's for you, you're going to have a fantastic life. But, and if it's not for you, that's fine, you'll have a fantastic life. Not doing it. But, but don't do it because your friend did it and made a lot of money. That is the wrong reason to do it. In my journey, what I always loved doing was solving pain points, solving real problems. And what I see in not just tech, but many, many entrepreneurial journeys is they've got an idea and they hang onto this idea and they find other people and they raise money. They do all that and they keep hanging onto the idea, even if the idea does not necessarily close the gap on some enough people's pain points. Right? Yeah. And so, you know, it's kind of the solution with no problem. And we make lots of solutions that don't have a problem. And we see this in tech all the time. We see in consumer products, we see in dumb tech devices like the juicer that squeezed a bag to make, you know, they lost $400 million or something. When people realize you could just buy the bag and squeeze it yourself. Right. These are like, nobody was actually looking for a bag squeezing juicer for your counter. Like nobody needed that. Right. It's not solving a pain point. So I, you know, I think some maturity in entrepreneurship is that my ideas probably aren't right. But, but, but you're going to have better ideas and better solutions. If by walking down the street you see a real pain point, like people are in pain with this. And then there's another trick here. It can't just be anyone in pain. It's gotta be people with money are in pain. I hate to say it, because if you make a product for people who make a dollar a year, they can't afford very much. It's a, you know, they're gonna give you a penny or two. But if you make a solution that really has an ROI of thousands of dollars or even millions of dollars a year, and AI can do that in some, some sectors now, people will beat a path to your door. Right? Because they can see if they spend a million with you, they save 10 million over here. This is a good thing. And I mean enterprises. So look, I've, I've killed companies, I've sold companies, I've bought companies, I've, I've had successful companies. I've got 94 worldwide patents just applying for a 95th right now. I keep inventing. I have teams around me that invent with me. It's certainly not just me. And I have fun and that brings me great joy. But the number one joy is when you've come up with something to solve a person's pain, company's pain, whatever, and they buy it and it solves their pain. That is a joyous moment.

I agree. I really love it. And I think one of the things that you really do exemplify here is it's not just solving one pain point, but because you've been involved for much longer Talking about an AI 10 years ago, you've been in the arena and it's not just going, oh, look, that may be a problem, but you're actually engaging on a consistent basis and reiterating and adapting as you continuously expand into that space.

Yeah, you have to. I started in the AI space inventing the first virtual assistant, which was, her name was Mary and it went under the name of Portico in my Talk and Magic Talk and eventually GM OnStar, Virtual Advisor. Eventually all of that technology that we developed got licensed for Siri and Cortana and Alexa and all of the other virtual assistants as you know them today. And that work was done in the mid to late 90s at a company called General Magic. And, and she, Mary was amazing. She could answer your phone and if, if, if she recognized that you were in my contact list, she would make an appointment on my calendar for you, you know, automatically, if that's the rules I set up. And she sounded human. She was a recorded human. Mary Mack was actually her actual name. So we started there. And of course that was a lot of rules at the time. A bit much more rules based engine and some AI. But over time, the accessibility of AI and machine learning has gotten better and better, mostly because of cloud and intense increases in CPU and GPU horsepower. Without the horsepower, we couldn't have done it right. So now we've got the horsepower. It's like, I think most people think that these large language models, ChatGPT just fell out of the sky one day. We've been working that for 70 or 80 years. Right. And making strides and it's not that someone couldn't have thought up the algorithms that we use today. Fifty years ago, they could have. We had no compute horsepower to do it. Right. The largest computers at the time, even in the 70s, would have taken a thousand years to process what we can now do in minutes or seconds. So you would have never trained a model on, say, a trillion tokens? That'd be impossible. There's nowhere to store it. But now with the cloud and GPUs as fast as they are, et cetera, we can consider that we could train a model with a trillion tokens. So this is a really fascinating time and really for entrepreneurs that don't have to be technical either, because you can look at the advent of large language models, just as we did with Excel and math models, you know, 50 years ago, and say, what kind of company can I create leveraging that technology? You don't have to go build a foundational model, but how can I use it to come up with something that really solves someone's pain and has enough of a moat around it where you aren't going to have a hundred other companies do it? So it's a fascinating time. Best time to be alive.

I fully agree. I think it's. Yeah, it's. Everyone's like, oh, I'm worried. I'm worried of what's going to happen, all this. It's like, yes, but what we're going to be capable of, yes. Humans still have to deal with being humans, and we still have to kind of find ways to look for the greater good, but looking at what's possible, the medical advances, the environmental, that hopefully we'll be able to take care of. Actually, when going through a lot of your material, you get asked a lot this question, like, well, what is AI going to do to my job? Is it going to take over and all that? I mean, let's not talk about that, but let's talk about from an entrepreneurial view, I hear a lot like, oh, AI is going to take away this type of business model or that type of business model. How do you see AI impacting entrepreneurs moving forward?

Well, many ways. So. So first of all, some of our costs as entrepreneurs had to do with generating marketing content and, and generating advertising, generating blog posts, and summarizing podcasts and all kinds of things. Right? Okay. So it's not that we haven't had AI to do that in the past, but large language models now make this approachable with no coding. With no coding. Not everyone's a coder. That means you can walk up to a large language model of chat, GPT or bard or, or whatever, and essentially talk to it in your own language and say, this is what I want. I want a blog post of 500 words or less that blah, blah, blah, blah, blah. And here's the, here's the information to learn from. And it needs to be in this tone. And 30 seconds later you have it. And you can edit it and you can clarify it and you can read whatever you want to do. And 10 minutes later you posted it. Now that's a post that might have taken you two days, including some writer's block in there, right?

Exactly.

And all of a sudden I'm done. I don't know. In 10 minutes. This is amazing because that means the entrepreneur who at the beginning has to do everything, him or herself, right. All of a sudden can collapse some very important tasks that would have taken days out of their schedule, collapse them into 10 minutes. Okay, I need more models that do that right in everything I do. Now let me give you some other examples. You say, I need a business plan for the following idea about blah, blah, blah, blah, blah, Boom. Business plan. I'm not saying it's right or wrong. I'm saying there's a plan. Okay, I need a test plan for when I come out with this product. How should I test it? Boom. Here's 10 things you should do. This is amazing. We have never had an assistant next to us that knew everything and it knows everything. It also knows fiction. So sometimes it gives you junk. Yeah, but, but, but basically every single thing that you think about and do, ask the model all the way to including, I have to work an extra four hours tonight. What are, what are the best words to use with my spouse? So, so that he or she doesn't leave me this time. Okay. He is excellent at giving you that advice. So before you say anything, even before you write emails back, I just got an email from an investor and the investor says, I'm not interested in your idea because of X, Y and Z. What should I reply back? If I want to re engage them and get their money, it'll tell you and you can decide if you use it or don't use it. But it's only. You're going to invest 30 seconds to get another opinion. Right there.

It's funny when you talk about that, I've been mixing and matching different styles. Like crystal. Crystalnose.com is a great one because it creates personality profiles off of public writings with people. So it gives you like, oh, are they Going to respond to verbose, short, direct, whatever.

Yeah, yeah.

I have to say, it feels sort of like cheating, but like, my responses are much more positive when I filter my writing through that and then re edit based upon, oh, wait, no, they're too long, or, you know, this person wants direct bullet points or yeah, likely.

That that is absolutely correct. And, and you know, it's giving you an educated opinion that you didn't have. And also, we as humans tend to respond first emotionally, even if we don't want to. That is the, you know, anytime we're in an argument with someone, there's an argument because two people are arguing and that's because one is emotional and then the other one gets emotional and there's this whole loop and we're disconnected from our frontal lobe. And the next thing you know, two people hate each other for some reason. That could be husband and wife, could be work, whatever it is. Right. And so if you just stop and say, you know, before I respond to that, let me ask the model how I should respond every single time it comes back with, it's something as if your psychiatrist was right next to you going, aj, I know you, what you want to say, but if you say that, it's going to ruin your relationship for weeks or months or ever. Yeah, maybe here's a different approach. And you bite your tongue and say, well, boy, I'm having a hard time. But after you do it, you go, actually, this is pretty brilliant. It's going to get me the result I actually wanted. Yeah.

And that's, I think, where some of the fun as we keep looking at, because like right now there is switching costs, moving from one platform to the other, taking things around. But as things move forward, some of the plugins, they're still a little janky, but watching this keep coming further and smoother, um, I'm excited for that idea that like, all right, in normal life I work hard or whatever, but, you know, maybe 80% effective if I'm really good at things. Having that extra 5, 10% from, like you said, just like that may not be the best response for you may need a little bit deeper.

I.

That's exciting. And I just wonder, what are we going to be selling? What are we going to be promoting based upon it when AI can handle so much.

Sure. Well, let's look at productivity this way, right? If you've got a startup, you're an entrepreneur. Maybe it's you, maybe it's you and a few people, whatever it is, at the beginning, if you would not take A pencil and a ledger book and start to keep track of your finances at the business. And you don't do that because one, we have Excel or two, we have QuickBooks or the like. Right. So we have tools now that actually in that, in the task of math and finance and I'm giving, we'll use finance in particular, probably makes you and me 10 to 100 times more productive than we would be with a pencil and more accurate. Some number like that. Huge. Just in that one task. That's a task that could have taken days and now takes a half an hour, whatever it is. Right, very. So 10 to 100 times more productive. That is what large language models are actually doing for our language tasks. They're making us for that task 10 to 100 times more productive than we would have been. And we may or may not have better outcomes, but we certainly used our time better. So if we've now automated math and we've automated now language, including decision making. Right. You can ask it, what decision would you make? Blah, blah, blah. So we've now automated the two big things that any entrepreneur has to do. They got to do math associated things, and there's lots of those. And you have to do language associated things and there's lots of those. Right. Now there may be some other things. Obviously there's other things you do during the day. You have to, I don't know, take, take this to the bank or you have to take the dry cleaning to here, or you have to, you know, fire up the chemical mixer, whatever it is. Right. That AI is not going to do. But, but all of those other desk tasks.

Chop the wood.

Yeah. All these other desk tasks are getting automated in a way that helps you. You're still in charge. You, you know, you're the master of Excel. It didn't go off and run some math without you. Just like you're the master of the large language model. Same thing. You are the master of it. You own it. You tell it what you want, you tell it what to do and you take back and you either take it or leave it, or regenerate it or fix it or whatever, you're still the master. I think when people realize that, they go, oh, this is just an incredible tool set that finally automates the other half of my time at my desk, including emails. I can apply this to emails. And of course, in Outlook, we've already now got starting to be built in and everybody's copilot for Outlook, copilot for writing presentations. It's phenomenal. The Presentations look better than I could do. And I can also go out to Stable diffusion or Dall E3 or whatever and generate amazing images and boom, put them right there. There was an image I needed for one of the startups I'm working with in the home theater space, sort of. And I just needed this exact picture of these people watching this. And these are the speakers and this is what it looks like, the size of the screen. I just described that all. I could not find a photo anywhere. Even, you know, not that I would borrow a photo, but even, even, you.

Know, stock clip art. Yeah.

Nobody had taken one from that angle. I had one in 10 seconds. There it was, boom. Put it right in the presentation. It looked like I shot the photo, except they didn't shoot the photo. This is, this is like I said, amazing time to be alive and an amazing time to be an entrepreneur. If you think of all the presentation you have to write, do make look great, I would do pitch decks, send them to an artist and we go back and forth for three weeks. Even though the pitch deck is going to change all the time right now I write the pitch deck and then I go, I need art for this, this, this, this and this. Boom, boom, boom, boom, boom. Done. I'm sorry about the artist, but I didn't need to do it. And by the way, give me a better layout for this. Okay, here. Boom.

Yeah.

Wow. This is amazing. Right? So this changes everything you do as an entrepreneur, including the testing of markets. You know, you can ask chat gbd, what do you think of this idea? You know, I want to make fireplace stones. I'm looking at the fireplace behind you. If that's a fireplace. I don't know that. Yep. Yeah, that, that look the following way, but go together automatically and it'll, it'll have some great recommendations and even know how to solve problems and problems with employees. You know, it's amazing.

It's funny you say that because yeah, I have been experimenting with some of those like Avatar Discovery or yeah, Avatar. You're like, okay, here, ask questions. And in the past I've invested and been involved with various user testing platforms and I did something and sent it to a friend who's an ex CEO of one of them, but he sold and he was like, oh, this is really good. Who did you use? And I was like, yeah, this was earlier. Now he's even like more of it. I was like, chat GPT. He's like, oh yeah, we had to go find like 10,000 people to go answer these questions.

Right? Exactly. It is an amazing time. And so that's why I think as an entrepreneur, you know, the only thing we don't have is time, because time is money. And we're burning up money as we take time. And your whole day goes by and you put 12 hours in, you go, I only got done with half the tasks. And okay, start using these accessible models that are now in English and you might find five hours of your day comes back to you.

That leads to a very good question I would like to ask you. Because of your involvement, the time you've been in the space and what you're doing now, you really are very much involved in a focus use of AI. I think a lot of entrepreneurs, we use the tools, you know, we have a very haphazard, especially when we're in that second inflection point into seven figures. We stopped rubbing sticks together as fast as possible, and now we're trying to figure out how to create structure and do this in a way we can actually sleep. That's my goal. Usually it's can I actually get to where I'm sleeping normally? But we throw tools at problems as we see them. We see whatever people are finding.

Cool.

Hear from podcast shows, whatever, where the tools are. How can we actually develop a practice around understanding? Because it changes so quickly. It's not about which tool is the best, but just being adaptable to AI. How can entrepreneurs start building an AI understanding or an AI practice?

Yeah, so I think there's two things, right? One, what are the best practices to use it within my company, me and my people, and where do we want to use it and where do we need to be cautious? And where can't we use public models? Where we have to have a private model or private access so that we're not giving certain things away. So, so I think you do have to put those best practices in. And best, we'll say models, for lack of a better word, because it isn't about ChatGPT versus Bard. It's really about often third part, you know, third party, other startups, right, who have, who have really done great embeddings or great unique uses of those models so that it's much more valuable and usable to you, right? And that's. And that's what you want to do. And so I just think, look, you have to take the time to find what works for you, what really adds value, what really adds productivity, and then launch that out throughout the rest of your team, however big that team is, and then encourage them to use it like no Coder should not be using either Codex or Copilot. It would be ridiculous. Although most coders aren't, but the ones that are 55% more productive than the ones that aren't. So please use it. You know, it doesn't work all the time, but you're going to double your productivity. That's pretty amazing, right? 55% more productive, and that's across like a million users. So I think that's fabulous. And then the other side of that is, are there places I can leverage this kind of technology to build out a product so that the technology is embedded in my product? Right. You're not going to go build a $7 billion foundational model that ChatGPT did, that OpenAI did, but you can leverage OpenAI and ChatGPT or other models or llama to whatever, in ways that are highly unique that you could build a real mode around. And it's certainly, you know, at Appvance. You mentioned Appvance. We have invented methods since the first product came out in 2018 to leverage AI and machine learning to find bugs in enterprise software. Where we are today is it. We'll find 10 times more bugs than your entire QA. I know what this is going to bring up. It brings up the question that we're not really going to talk about today probably is, oh, how about my job? Look, there are certain jobs that are going to certainly be impacted by AI. Most of those are not in the United States today. And that's simply because. Not because we're being picky about who gets laid off or who's. Because swaths of. Of certain kinds of jobs were sent offshore over the last 20 years. And those happen to be the things that were the most price sensitive. We could not afford to pay $20 an hour. We wanted to pay a dollar an hour, so we went offshore. We pay a dollar an hour. Tier 1 customer support. That's an example. Certainly software QA, and there are many, many others. Right. Those will be the first. They are already the first jobs that are under threat from AI. Again, not because they happen to be offshore, but because of the very nature of why we sent them offshore. We sent them offshore because they were relatively easy to train people and they were relatively repetitive tasks. And if I can train a repetitive task, then I can do those offshore. And I don't feel bad about that. It's a great use of human resources, and these are great people taking those jobs, but I don't need to do them onshore. I don't need those people. Right. So those are the first jobs that are under threat. And, and they're under significant threat over the next one, two, three years, not 20. It's like most tier one customer support will be a chatbot or will be a virtual assistant, period, full stop. Frankly, the machine does it better. It's more accurate, it gives more accurate results more often. And, and you can update basically what it says on a dime as we learn more. Right. As we change our FAQs or whatever. So look again, exciting time. If you're an entrepreneur and you had tier one customer support, whether it was here or there or whatever, really, you can have a chatbot do that a week from now. It's very easy. I won't, I don't think we'll get into embeddings and how that's done, but there are companies that do it for you. And now, frankly, OpenAI has basically sort of launched the ability to do it right on their platform. And you don't need necessarily pinecone and all these specialized databases. Right. So it's really fascinating. It's, it's just a fascinating time. And if you can cut your tier one customer support costs from a million a year, $20,000 a year to pay for the. Pay for the AI, or 30,000, that's incredible. That's more money back to you to drive the cost of your product down to be more competitive.

Yep. And that does seem, it's like you're going to be able to reach a larger market, go further because the cost to deliver the create is going to keep coming down until such point that pennies across larger and larger markets will theoretically give you a better business.

Yeah, I, it's automation since the beginning has driven product cost down. And I'll give you an example. Cars. General Motors put in the first robot in 1961. The first robot in 1960. Today they employ more people than they did in 1961, yet they've been rolling out robots all the way since 1961. And you know, GM has hundreds of thousands of robots today doing all of these kinds of tasks. And people say, well, I don't, I, you know, I don't understand because cars are so expensive this and that. If you did not have any robots today, a car would cost you $300,000 to be man made. A basic car. About 300,000. Yeah. 250. 300,000. So by, by putting robots in, they have driven down, they and Toyota and everybody else have driven down the, the sort of average cost of a vehicle compared to what it would have been had they been made manually. Right. This is a huge deal. And that allowed more access to more vehicles and motor vehicles for more of the population of the world by having a $20,000 car instead of a $200,000 or $300,000 car. And this is true all the time. Automation drives down the cost of goods and services, which increases the interest in those services because more people can afford them. Right. So it increases your business, which ultimately in an entire country, increases your gdp. So the most productive countries, of which the United States is pretty much the top of the line, the most productive countries in terms of output per worker, tend to also have the highest GDP and the highest standard of living. And yes, I know that doesn't. You know, there's people and you know, this isn't a political podcast, but you get the point. Right? So.

Yeah.

And I'd much rather live in a country and be part of a. An entrepreneurial culture that encourages further and further and further productivity increases that drive up the revenue and drive down the cost of goods and services. This is absolutely amazing.

Yeah. And it's also, you know, there's a lot of, oh, productivity at all costs. It's like. It's not productivity. It's also driving quality of life. Because as you said, who really wants to be doing basic structural work?

Yeah.

Even if we're just talking white collar, that basic drug work. It's not hours upon hours of creating the template structure so your boss could have a presentation.

Right.

As someone who's gone up through that, it's like hours upon hours of nothing.

Real.

Yeah.

No.

Real value generation.

No. Who, who would want to sit in an office today and say, I am ready. I have my pencil and I am ready to do long division. I have my pencil. You wouldn't hire someone who does that. Why? Because a machine has done that for 50 years. A calculator and then Excel. I mean, we don't think about long division anymore. In fact, you don't really need to learn long division. It is not a useful skill. And I know people, people always yell at me when I say that, well, the kids need to learn, and it's okay, great. They are learning a skill they will never need again. But if you want to teach them that so they have the understanding of how the math works underneath, great. But let's not push it much further than that. And this is, this is going to be true with writing. You know, this is a big debate. I'm on the board of rit, Rochester Institute of Technology, and it's a big debate. There and at all universities today is the instructors mostly say you cannot use under any circumstances a large language model like ChatGPT.

Yeah.

And the kids walk out of the class and they run a chatbot and does their homework. Is this a secret? Like, like we're shocked. I know. We used to have to pay someone to do it. Now the thing does it for free. This is amazing. So, so I think instead what we should say is the skill of writing a paper we thought in the work environment would really be needed and was for a long time you had to write business plans and documents and papers and stories and case studies and whatever it was. Right. However, going forward, when I'm hiring people, I don't care if they can write because they like I have a machine that writes for them now. So I do care if they know how to prompt and I do care if they know how to edit. In fact, if they said I'd like to write it by hand, I'd say, I don't want to hire you. You're the wrong hire. So what we need to be teaching in our universities now is about one minute of write this one in my class by hand. I just want to see what you do after that. I just want you to use ChatGPT because you have to be the master of that tool or you will not get hired.

It's so funny. My son is in his first year at St. Andrews over in Scotland. One professor would say, look, the expectation of what you deliver, I've increased. I now expect you to be able to a more well developed thought process. A more developed argument versus this says you're going to use, you know, you may or may not be using this, but I expect now, compared to last year.

Yep.

That I want to see a higher level because you'll have the resources to go deeper into the argument. And I thought I was like, that's brilliant. That's exactly what will be more helpful for my son going forward than okay, I didn't use it. But yeah, I really just edited.

Do you remember in our day it was, you know, now when you go home to do your homework, you are not allowed to use the calculator. What I. Did you see that long division problem? This is dumb. Right? And, and, and, and so of course you'd have show me your work, you'd have to show the long division. But, but the fact of the matter is, is I never had to do long division again in my life because we had a calculator. So. So it is not useful to teach students things that they will never use again and again. I know there's big arguments for underlying, but underlying very quickly. Then move on and use the tools at your disposal. You'll be expected to do so. And when the wheel came to town. It's one of the things I say in my keynote speech. You know, we've been scared of new technology for thousands, a hundred thousand years. When the wheel showed up, the guy who used to carry the goods up the hill looked over and said, I'm in real trouble here. Like, my whole job is over. Except some guy said, my job's over. Went home and said, I have no more life. No one will ever pay me to carry their goods up the hill. Other guys who used to carry the goods up the hill ran over, got two wheels, built a cart out of it and carried five times more. And all of a sudden that's. So those who embraced it went on. Those who didn't embrace it starved. So this is one of those moments.

And I agree. And what's funny is, even with that example, there is such a great thing because was it the Incas? They had toys with wheels. Yeah. They were in places very mountainous. Whatever. They had it. And the concept of the wheel and moving things, toy level on wheels was something they understood, but never large objects. And it was just like not used that fascinating. Like, oh, that's cute little toy. Yeah.

Didn't put two and two together, but someone eventually did. What? Someone said, there is. I see this problem over here. The problem is we're carrying all these things, like on sleds and just dragging them along. Wonder what would happen if I put bigger wheels on the sled. Someone did that one day and go, oh, this moves a lot better. It's a good idea.

And came over with germs. Yeah. Some people who had done it elsewhere and had germs and a few other things and steel, that minor thing, that minor thing that changed course of things. Well, earlier you said just the ability to solve problems has driven you. How would you go back to finding success for yourself? Because it is so hard as an individual, as an entrepreneur, to one celebrate success because success is just a number on the line, While failures are daily occurrences, it feels. How do you go about building a practice of celebrating your success and then defining in the long term what that success will look like?

Well, it's a great question. I think all of us define our success at some level by having a successful outcome, a successful company, a sale of a company, all those things. Right?

Yeah.

Success for our investors, et CETERA but success probably starts, you know, with solving those pain points. And that starts with curiosity, right? I'm curious about the world. I'm curious about what makes that tick. I'm curious about why that is a, why is that a problem? I saw, I saw that today. Someone brought me something and said, and why is that even a problem? How did, didn't, didn't anyone see that that needed to be solved for the last hundred years, right? It's like, how can I. And then you get curious and you say, what, what did other people do? And you go read patents and you see what people did and you say, I think I can do this better. Like, there's a real need here. People have a real pain. You know, when we invented soundproof drywall of, of all the crazy inventions, right? It was because there were builders that were getting sued building condos and townhomes and, and everybody could hear what was going on next door, and then they'd sue the builder. And the builder said, I'm, I'm using the best practices. And I looked at the best practices and said, the best practice is who invented this? Whoever invented this was an idiot. It was a, it was a, it was a technology that never worked well. It didn't work well since they invented it in the 60s. It could easily be acoustically shorted out just by hanging a picture on the wall. It was a dumb idea. And so I said, well, I wonder if one could make drywall itself that acoustically changed acoustic energy into heat energy. A minor amount, that's immeasurable, but nevertheless so that less energy would pass through the wall, right? And if I could do that on both sides of the wall and construct the cavity correctly, I could probably reduce the sound by 15 or 20 decibels. And that's what we did. And it took time, obviously. And then people thought we were crazy. And every drywall contractor told me I was, who's going to buy a $50 piece of drywall when regular drywall is $10? I said, people who don't want to get sued. And it's only on the party walls. So, you know, this is a really good idea. Oh, no one will ever do it. Well, you know, today that is the standard way that you build party walls and high end hotels, as well as multifamily and obviously recording studios and things like that, right? And there's various levels of it, but it costs more. However, the labor isn't anymore because it goes up with the labor of one piece of drywall. So all the cost in Your drywall hanging, by the way, is the labor. Right. So drywall doesn't cost $10 to hang. It's $10 for the drywall and upwards of 40 or $50 for the labor. So if the drywall goes up to $50, but it's still $50 for the labor, you doubled the cost of that wall, but it didn't go up 5 or 10x. Right. And so again, I was curious as to why no one had ever solved this. I'm still curious to this day. But we did solve it, and it became a very big business and we sold that. But again, pain point. Right. So how do you describe your success? So I'll tell you, every time I see someone using Quiet Rock or a truck goes by with Quiet Rock on it, I feel immense pride. It's not about the money, it's just immense pride. I did that. Right. I solved this problem. For now. Millions of people. We retrofitted the windows in the Empire State Building with high R value windows that we developed. And they are three to four times more insulative than dual pane. And they don't have more glass in them that uses suspended film and different gases and the net and they don't weigh any more than dual pane. And the net result is the Empire State Building became a brand new building with brand new glass, brand new windows. We reused the glass, cleaned it on the fifth floor, put it back in, and now people could sit next to the window and wanted to sit next to the window because it wasn't cold in the winter and it wasn't hot in the summer. Immense joy. Right? If you're an entrepreneur and people are using your product in that way, it brings immense joy. And you want joy in your life. And so I'd say, you know, yearn for the joy. And yes, money's good. Absolutely. I hope you get paid. You don't always get paid for your inventions. You don't always go down the, you know, you're not always successful. But when people are using the thing that you develop to solve their problem, and whether it costs more or doesn't, and they come to you and they say, this really solved the problem. Thank you, Kevin. Okay, that's cool. Everyone who uses Siri or Alexa or any of that today, same thing. It is based on the work we did at General Magic. And when I use Siri, I go, oh, this is really cool. I came up with this.

I know this.

Yeah, yeah. So. So lots of pride in these things. And I think that's certainly true. Advance we're working on AI that, that does find bugs in software. And every client of App Vance that, that, that I talk to and it says, oh, oh, my goodness, not only did you find 10 times more bugs, but you found this one that could have just really killed us. Thank you. And it wasn't me, it was the machine. Right. But the machine was smart enough to find bugs that humans couldn't find, but would eventually find. Cyber security ring. Token Ring. Another company in Rochester, New York, that, that we helped to restart, and it's doing great, and it's a cyber security ring. So everyone, you know what two FA and MFA are? Yeah, they, they, they are numbers that come on your phone or an authenticator app or whatever. What most people don't know, but now they're about to in this, in this podcast, is that virtually every single ransomware hack over the past year has also hacked an MFA to get to the systems. All legacy MFA has been hacked, period, full stop. And it's hacked every day, every hour, every minute. And it's become increasingly easy to do. So there are technical hacks and there are social engineering hacks for all of them, and they all work every single time. And I'll give you one stupid, simple social engineering hack, because I think people will enjoy this. Okay, you've got a Microsoft authenticator app or something on your phone. And you know when you. Sometimes you go to log into Outlook or something and it goes off, is that you? Is that you? All those things, right? So you say, yes, it's me, and you confirm it on your phone and your device and all that. Okay, 1:30am in the morning, that goes off, Is it you? Is it you? Is it you? You know your Outlook app is trying to log in. Is it you? Is it you? Is it you? Okay, if I do that to a thousand employees in a company, I'll get a hundred by 2am says, yes, it's me. I can't stand it. I can't turn it off. It's probably my application trying to log in or trying to update. It's.

Yeah, some remote.

Yeah. Okay. So I didn't even have to break into the app or hack the app. All I had to do was what's called an MFA fatigue attack. I know you're fatigued with all this MFA stuff, so I'm going to play that to my advantage. So since a hundred out of a thousand will do it, it's very easy for me to get into any company's system. I just need to go out to a thousand and in an hour I'll have a hundred who give me their mfa. And you go, well, what about the login ID and password? I already got that off the dark web, so I don't even need that anymore. So I just gave you the simplest of hacks. So Token Ring is a next generation mfa. It's the only real one right now. It's wearable. It's a ring. You put it on, it's tied to your fingerprint, your fingerprint. Only if you lose the ring, it doesn't work for anyone else. Nobody gets in. And you can't hack it with any social engineering tricks because the ring doesn't have a readout on it. It only reads the backend server directly, 128 bit encryption. So there's nothing you can do. Like if a hacker called you and said, give me your code, you'll go, what code? The ring doesn't have a code on it. Right. It talks to the server.

Cut my finger off.

Yes. They could cut your arm off and take your arm and flop it around.

It's a bad movie. Yeah.

Generally they're in Russia or two, China, so they're not likely to do that. Right. Anyway, so Token Ring is a great example of next generation mfa. We see two years ago that that legacy mfa, the minute we started rolling it out, was so easy to hack. And so we needed a next generation MFA that's unhackable and that's the unhackable solution.

I mean, every night this weekend, yeah. I wake up in the morning and there's some prompt at between 1 and 3 in the morning, some email or prompt on my phone saying, oh, are you logging in? Yeah. Please confirm you're logging in to your Google account.

Yes. There you go. And it's probably not you.

No, I sound asleep. Kevin, I cannot thank you enough. This was so cool. How can the audience learn more about you, your LinkedIn, what's the best way they can learn?

My LinkedIn is case to race. But also you can go to my website, case kevinserrace.com easy to remember, my first name, last name, dot com. And it has all kinds of updates and information and my LinkedIn. And I do answer my LinkedIn messages if people want to message me. And I'm out there doing 30, 40 keynotes a year. So you can go to my website and go to any of the booking agencies there, book me for a keynote. So we have a lot of fun talking about AI. I talk about disruptive innovation. I Silicon Valley, disruptive innovation. There's so much to talk about, how companies get started, how you get funded, you know, what the risks are. All, all, all about entrepreneurship. So these are all great topics and we try to entertain the crowds too.

Yeah, no, I mean, I could definitely understand the entertainment and the education altogether. I mean, this is truly entertainment opportunity and we barely scratch the surface of your background. I mean, this has been a fascinating. Kevin, thank you so much. I'll make sure I put everything in the show notes and our socials and email when we go out. But Kevin, thank you so much for coming on the show. I really appreciate it today.

Thanks for having me, AJ.

Hey, everyone, thank you so much for listening today with Kevin. Go check out his site, go to LinkedIn. I have so much more you can dive into about the entrepreneurial journey by just going through his material. So much is happening and I think playing around and using his example of exploring problem solving and using these tools, AI and so much more, it's going to take you much further on your own entrepreneurial efforts. So, hey, thank you. Can't wait to talk to you again. Have a wonderful day.

Bye.

Bye.