Matthew has more than 20 years of experience in starting and running high-growth tech companies, especially in the areas of machine learning, predictive analytics, and decision optimisation. He is currently the CEO of Complexica, a provider of Artificial Intelligence software for supply & demand optimisation, and a director of several ASX-listed companies.
In today's podcast, Matt and Daniel discuss Matthew's book in which he co-authors, called The Rise of Artificial Intelligence. Matt's book is important for any modern business leader, executive or manager, who is interested in the learning the basic fundamentals of Artificial Intelligence and what propelled to the level of popularity and prominence that it enjoys today. Matt also talked about the potentially existential threat of AI, how he is much more optimistic about the rise and benefits of AI, and also shared some concerns and provides with a different perspective on some of the risks with the Rise of AI.
Where to find Matthew Michalewicz:
LinkedIn Profile
Website: https://www.michalewicz.com.au/
Complexica: https://www.complexica.com/
Download the first 2 Chapters of The Rise of Artificial Intelligence book here.
Join the conversation on Synergy IQ LinkedIn, Facebook and Instagram (@synergyiq) and please support other leaders by liking, subscribing and sharing this podcast.
Access SynergyIQ Website to get to know more about us.
Say hello to our host Daniel on LinkedIn.
Matthew has more than 20 years of experience in starting and running high-growth tech companies, especially in the areas of machine learning, predictive analytics, and decision optimisation. He is currently the CEO of Complexica, a provider of Artificial Intelligence software for supply & demand optimisation, and a director of several ASX-listed companies.
In today's podcast, Matt and Daniel discuss Matthew's book in which he co-authors, called The Rise of Artificial Intelligence. Matt's book is important for any modern business leader, executive or manager, who is interested in the learning the basic fundamentals of Artificial Intelligence and what propelled to the level of popularity and prominence that it enjoys today. Matt also talked about the potentially existential threat of AI, how he is much more optimistic about the rise and benefits of AI, and also shared some concerns and provides with a different perspective on some of the risks with the Rise of AI.
Where to find Matthew Michalewicz:
LinkedIn Profile
Website: https://www.michalewicz.com.au/
Complexica: https://www.complexica.com/
Download the first 2 Chapters of The Rise of Artificial Intelligence book here.
Join the conversation on Synergy IQ LinkedIn, Facebook and Instagram (@synergyiq) and please support other leaders by liking, subscribing and sharing this podcast.
Access SynergyIQ Website to get to know more about us.
Say hello to our host Daniel on LinkedIn.
Welcome to creating synergy where we explore what it takes to transform. We are powered by Synergy IQ. Our mission is to help leaders create world class businesses where people are safe, valued, inspired and fulfilled. We can only do this with our amazing community. So thank you for listening.
Daniel Franco:Hey there synergises Welcome back to another episode of The creating synergy podcast. My name is Daniel Franco, your host and today we welcome back by popular demand, the great Matthew Michalewicz. Matthew is more than 20 years of experience in starting and running high growth tech companies, especially in the areas of machine learning. Predictive Analytics and decision optimization is currently the CEO of Complexica, a provider of artificial intelligence software for the supply and demand of optimization, and a director of several ASX listed companies. In today's podcast, Matt and I discuss his new book in which he co authors called the rise of artificial intelligence. Matt's book is important for any modern business later, executive or manager who is interested in learning the basic fundamentals of artificial intelligence and what has propelled it to the level of popularity and prominence that it enjoys today. There was no real topic too difficult for Matt, as we talked about the potential of existential threat of AI, and how Matt is much more optimistic about the rise and benefits of AI in lieu of a company like Skynet, taking over the world like we see on the movie Terminator. Matt also shares some of his concerns and provides a different perspective on some of the risks that the rise of AI bring. That's been kind enough to share the first two chapters of his book. And there you will see them attached to our show notes. This is a jam packed episode food with knowledge that has been extensively researched, and one that I know everyone will enjoy as much as I did. Hope you check it out. So welcome back to the creating Synergy podcast. My name is Daniel Franco, your host, and today we welcome back. Matt Michalewicz. Now I pronounced that wrong last time.
Matthew Michalewicz:Yeahyou pronounce it wrong this time?
Daniel Franco:I just remembered as I said, How do you pronounce your surname
Matthew Michalewicz:Michalewicz
Daniel Franco:Michael which There you go. So you are packed back by popular demand, we have had some really, really great feedback. One of our most listened to podcast, I've gotten text message after text message from entrepreneur after to anyone who's really in business and wanting to grow their career. messaging me telling me how great that was to listen to people listening to two or three times. Wow,
Matthew Michalewicz:most people don't have me back. So that's very nice.
Daniel Franco:Nice. Great to have you back. On the podcast. We talked about getting you on for the discussion in AI, because you mentioned your writing Yes. Book. Yes. Yes. Can you tell us a little bit about that book?
Matthew Michalewicz:Yeah, so the name of it is the rise of AI of artificial intelligence. It's focused really on business applications to improve revenue margin outcomes. So it's written for a business manager, an executive, someone that doesn't come from a technical background, and really wants a demystification of the subject, starting with an introduction to the field, what is it all the way through at the very end case studies, application areas and everything in between prediction, machine learning data, etc, all the all those elements you really wanted to understand, but somehow rather have never gotten around to kind of piecing all the all of it together to get a view of it. So very much oriented towards business people.
Daniel Franco:So I'll definitely get a copy of that first copy.
Matthew Michalewicz:I've already promised the first so you have the second.
Daniel Franco:I'll take it. So what is your definition of AI? I mean, we all know that AI is out there. It's changing the world. But what's your definition of AI? Yeah,
Matthew Michalewicz:my definition is it's a field research field that tries to replicate humanity. That's my definition of it. It is comprised of a lot of subfields and sub research areas, which we can discuss in in this podcast. But overall, it's an attempt to artificially replicate ourselves as as a view so to be artificially intelligent, you typically need sensory information as well. So you need to replicate site you need to replicate touch, you need to replicate movement so you can sense the environment you need to replicate communication speech hearing, and what you end up actually replicating is the entire human biological body. It's a fascinating area. And and quite old and dated, despite the recent hype, which I find the most interesting.
Daniel Franco:Well, yes, we'll go on to that was Alan Turing, the sort of Was Yeah, is he the mastermind or the inventor behind
Matthew Michalewicz:many people credit him as being the father of artificial intelligence, I think he had the first public lectures and publications where he talked about thinking machines and, and machines that can learn by themselves from experience and so forth. And, and I think that was the late 40s. from a long time ago, it
Daniel Franco:was a long time ago, which is amazing considering where we are today. So is the ultimate idea of AI to replace humans in work?
Matthew Michalewicz:It depends on who you ask. If you ask the worker, no, if you ask the CEO that is of a major corporation that is incentivized to create labor efficiencies and improve bottom line, and is bonused on shareholder value and returns. Absolutely. So and this is an interesting point that you've actually raised, I don't really believe that innovation happens just for the sake of innovation, I think it's really driven by capitalism. So you have this huge competitive forces and pressures in the market of all these public companies and large private companies competing with one another, wanting to create competitive advantage wanting to be more efficient, wanting to be more profitable. And they turn to technology to accomplish that. And this is kind of it's kind of like an arms race, except in business. It's an innovation race. So all of these innovations that are happening in the world are in large part driven by capitalism, to improve a competitive position of a company, and the God the ultimate dream would be to have, you know, no workers in a business, everything's done by machines, but then everyone would be out of work, and there'd be no one, no one would be able to afford the products of that company. So yeah, has to be a balance somewhere,
Daniel Franco:does what the diff what's the difference between machine learning and deep learning and artificial intelligence? Yeah, I guess, is there a difference? Maybe,
Matthew Michalewicz:yes, there is a difference. And, and I'll put them all into perspective, maybe I'll take one step back, and kind of in a minute, explain how all the pieces fit together. So So if we think of AI as replicating the human body or attempting to replicate the human body, there are really four predominant research areas. One of them is robotics, which is the replication of movement. and other one is computer vision, which is the replication of sight and image recognition, looking at moving imagery, videos, photographs, and so on. The third area is natural language processing. Siri, Google, Alexa is the replication of the speech, element of existence. And the last one, which is the most interesting and the most challenging is cognitive computing, which is the attempt to replicate brain functions. So you've got these problem areas, and then think of them as cutting vertically down in this four of them. And then cutting across these problem areas as you have algorithmic research into the various algorithmic techniques and methods. And they cut across because a technique like neural networks, which you can we can talk about a little bit later on, is applied in all four of those areas. So you have some computer scientists that specialize in a domain, they become scientists that specialize in computer vision, they experiment with a wide variety of different algorithms to see if they can produce better results on recognizing objects, etc. And then you've got other types of scientists, my father fits into this category that have been algorithmic specialists, and they specialize in an algorithm. So if you have this and the algorithms cut horizontally, so if you have that, in your mind, the problems cutting vertically in the algorithms cutting horizontally, machine learning is a set of algorithms that sits in that horizontal cross section. And within machine learning is a technique called Deep Learning, okay, which also cuts cuts across and so deep learning has been applied to robotics, division systems, natural language processing and cognitive computing. It's it's an it's an advanced form of neural networks, which has been around for a long time. Very good. And your father has just been awarded well recognized to be one of the most influential in the AI space is our career. My father's went to a lot of awards. Over the years, I was actually lucky enough to be raised inside of a university, so to speak, my father moved into AI in 1982. And I was six at the time and every day after school. So this was grade one, I would go to the university, which was like 10 minutes walking distance, sit in my father's office and play video games or draw pictures on the whiteboard. I think it was a chalkboard, I'm actually actually going going back to tidy to while in the background, he met with other professors or with PhD students talking about neural networks or machine learning or elements of the Turing test and so on. So it was from a very early age, kind of the vernacular that that I grew up with. And since that time, I think my father's written, gosh, three 400 research publications, dozens of books, and has become one of the leaders in an area of algorithmic technology called evolutionary computation, which he often gets, I think he's the most cited person in that area in the world and gets often awards and speaks at events because of an
Daniel Franco:obviously a good business partner. Yeah, to have within the complexity.
Matthew Michalewicz:You know, I think, actually, I'm not sure if we touched upon this before. But I, for the entrepreneurs that are listening to this, I always give the advice team up with people that are good at doing what you're not good at doing. Yeah. So there's probably nothing worse than three identical people with the same skills, the same everything teaming up and creating a business, that it will fail, but it makes it harder. So complex occur, and before it, solve it. And before it, new tech, all of those companies had really three very different people. So I come from an angle of sales, entrepreneurship, vision leadership, my father comes with the element of science. And then the third business partner Constantine comes purely from this enterprise software element to architecting, large scale software systems integrating, making sure you have you know, 99%, uptime, and so on. So the different skills make for a great partnership to be able to execute in the marketplace. If there was 3 million here, we wouldn't get it done. Yeah. You wouldn't be talking to me about it. Yeah.
Daniel Franco:He wouldn't be on my interview list.
Matthew Michalewicz:Thanks.
Daniel Franco:I'm steering the Michelle and my business partner, Michelle and I exactly the same, I'm probably sitting the same world as you a little bit more. On the sales side, the business growth side, and Michelle's very much around the leadership culture, building a sustainable business, and also the technical, she brings the technical experience to the service that was very important. Absolutely. Yeah, we need.
Matthew Michalewicz:I used to teach for a long time at Adelaide uni probably for a decade or semester course every year on entrepreneurship for computer science graduates. And I've, I've run it every year at complexica for the team over here and in the other offices. So I think, and that's the thing I always stress, get together with people that compliment you. Because when you think about all the things that an entrepreneur needs to do, to start and grow a business guys a lot of required skills. So if you can have a team where you cover the majority of those skills, not as you as an individual but as a team, but you have such an advantage.
Daniel Franco:absolutely, which is on the last podcast with with Schwarzenegger and how he said t at he built a team. Well, h's always all the success to is team, not his individu
Matthew Michalewicz:I thought it was just hard work in the gym. Yeah.
Daniel Franco:So just on complexica, can you offer AI service? I understand. It's Larry, the digital analyst. What if simulator, and the decision cloud is that the three
Matthew Michalewicz:I'll explain it in the easiest language. And it's software software that improves decision making in companies. And there's two elements to the software. There's the software itself that a user would use, they log in somewhere they see a screen they see swatting boards or planning boards, etc.
Daniel Franco:A lot. Larry has a seat in the has his own seat in the business,
Matthew Michalewicz:he has his own seat in the software. So what ends up happening is in that environment called decision cloud, which has modules to it depending on what business function you're sitting in, sales, marketing, supply chain, and so on. There are very complex tasks that need to be executed. And by asking Larry, you trigger the AI engine and through a sometimes it takes 10 seconds, sometimes a minute, depending on the complexity of the problem. The recommended answer can use there's two levels to it. The engine by itself would be difficult to use because there's no screens that you don't have access to. And then a system without any intelligence behind it. It's very nice. But God wouldn't it be great to push a button and have a machine tell you what to do based upon considering everything that's out there?
Daniel Franco:Yeah. So is that helping the CIA and the CFO and the likes to make decisions and the board to make decisions?
Matthew Michalewicz:Not so much at the board it's usually line of business. An example from South Australia's anyone that shops in sippin saves or celebrations or thirsty cat all of those stores if you walk in every product and the price and what that is in the catalog comes from our software as an example. So as you can imagine, based on what information was there, when the in head office in Sydney when these companies come up with promotions going forward and what pricing. They're really trying to make decisions that maximize volume and margin. And all of those decisions are based on prediction. So say, we were running the vodka category, the spirits category in one of these retailers, we might think that in December this year, it would be a good idea to put Absolut Vodka on promotion in South Australia 20% off. So to evaluate that decision, you need to predict how much more yourself which is time specific, product specific and geography specific, then you need to predict what kind of drop you're having sales from competing products, because people switch to the one that's on promotion. Yeah. And then you need to predict if there's going to be a basket effective, are people going to buy other products with the vodka? Or are you just gonna have the heavy drinkers come out of the woods and stock up in leap? So those are very complex predictions to make. And that's, you know, one product in one geography in one time period. Imagine trying to make 10,000 decisions like that. So and then the prices can be infinite doesn't have to be 20% off can be 21 25 50 241 341. Yeah, multibuy, etc. So yeah, so AI, the tying back to the theme is really well suited for complex environments, the more complexity exists in any process, the more difficult it becomes for humans to stay on top of it and make optimized decisions. Look at computer vision is used in security settings, airports, military facilities, stadiums, where you're trying to what have you ever seen on TV? We're in a movie where you have the guards sitting in front of monitors yet. Imagine if there was 1000 cameras? Yeah, how many people would you need to sit and stay on top of that to look for suspicious activity for someone to follow that's greatly suited for algorithmic technology, you train the algorithms on what suspicious activity looks like. And then can monitor all of those combined cameras and spot the things that you would be looking for if you had 1000 people constantly looking at each individual camera. So there's this huge relationship between complexity and the value that AI can bring to the table.
Daniel Franco:So on that using the cameras, as an example, no, no, it was only an example. But yeah, this privacy become a problem. When you've got 1000 cameras on you.
Matthew Michalewicz:It depends which country you're in. Yeah. In China, it's no problem. In Australia, it's a problem. It's a problem.
Daniel Franco:So is there a regulator or a governing body or someone especially in Australia, look, we're in Australia, that that is looking after this and making sure that people aren't creating the next godlike creature?
Matthew Michalewicz:I think there's two parts of that question. There's definitely rules and regulations around privacy and and what we can't and can't do, because you don't need AI, to, for example, collect photographs of people still imagery from different types of cameras, and then try to identify, you could do these kinds of things, in many ways. So so there's laws regulating what we can and can't do. On the flip side, I don't think that there is any governing body that has legal authority to curtail or control what is done algorithmically, or what isn't done algorithmically. You also raised an interesting point around privacy and my response around country. I had a conversation a few years ago, the person will run remain an unnamed, but he was the CEO of one of those automobile associations. So someone very close to the automotive sector. Yeah, he told me this very intelligent person that driverless cars will never happen, absolutely never happen. And he says, it's not a technological thing. It's what you're referring to Daniel, it's an acceptance kind of thing. And he says, The reason it will never happen is because they will never give a machine the decision on whether to kill the driver or kill a kid that jumped out on the street. That was the example that he used. And my counter, I said, is wrong for the following reason. Some your that comment is made from the perspective of an Australian value system. Yes. And there are so many countries in the world that don't have that kind of value system
Daniel Franco:at all.
Matthew Michalewicz:And those will be the places where driverless cars
Daniel Franco:where it starts and the point I'll have what she's having,
Matthew Michalewicz:yeah, but there'll be accidents. And yes, there'll be ethical decisions that have to be made, etc. But there'll be made in a cultural environment that has different values, as with any technology, you have improvement, and then once it reaches a certain level of improvement will be brought in to other countries in in limited use, so to speak. So I always view technology in that way. There's always going to be some place that will do something that you and I couldn't imagine doing. And that's because we're from a western country and have been raised in a certain way.
Daniel Franco:That's the the famous for a philosophical question. If you're, if there's a runaway train cart, yes, coming down the hill, and there's a split in the train tracks, and there's One train on one train track is a one person on the other train track there's five people. Yeah. And you've got you're sitting next to the lever. Yeah, currently the train tracks heading towards the five people. What do you do? It's it's Yeah, famous. Yeah, there is no right or wrong answer.
Matthew Michalewicz:I know that one person on the other well
Daniel Franco:it's one of those questions that there's so many What if now, you know if the if it's a group of elderly that a fireman a child on that's the one then that's a whole different scenario so it's it's really based like you said on the values correct of each individual and what they
Matthew Michalewicz:and I'll tell you one thing that I've learned in life is that those types of question even though I laugh about them, it's kind of a serious question. You know, it's, it's, it cuts to your core values as a human being and how you value life and so on. But I've discovered that whatever we say we would do in certain circumstances, we probably will sometimes surprise ourselves when you actually are in those situations. So the number of times that I have said, Oh, if I was ever in such and such situation, I would definitely do this. Yeah. And then I found myself in such and such situation. I didn't do that at all. I did something
Daniel Franco:You could question yourself
Matthew Michalewicz:you do something. So yeah. So one thing is logically looking at a situation and, and sitting at home with a, you know, a burgundy or something, cigar by the fireplace and having a philosophical one. And another one is being in the heat of the moment, absolutely emotion, the adrenaline, etc. And you got to make some real decision or no decision or no decision. And then you find upon reflection that what you thought you would do wasn't at all what actually happened in the real environment.
Daniel Franco:Brilliant. So there are three concerns of AI three main concern, yes. Those being the existential threat that it poses, you know, the rise of AI can wipe out the human civilization. There's the misuse, you know, like the Facebook's of the world using personal information, promotion, selling that whole sort of place. And then there's the overuse. I'm not sure if you ever seen the Disney movie the wall-e? Well, yeah, of course.
Matthew Michalewicz:They're all floating in space to clean up the garbage. Correct. And
Daniel Franco:essentially, what that does is it is giving rise to the useless class. Yeah, people become lazy and rely on technology. What is your thought process around? Probably the most worrying for you in those three categories?
Matthew Michalewicz:Okay, so all excellent questions. And, you know, without spending hours, let me give you my overall take, first of all, having been an AI for a long, long time, many people don't realize that AI has actually gone through hype cycles before this is this is just the latest. And those hype cycles have ended in disappointment. To the point where they even I think it was late 80s, early 90s, there was an AI nuclear winter, to the point that if you were an AI scientist, you wouldn't even use AI, because it was such a, you know, looked down upon a joke. Yeah. And and during that era, new terms came to be for example, soft computing became as kind of, or computational intelligence they began calling it CI to distance themselves from AI. Now all of a sudden, you know, it's hypee again, oh my god, you know, is it gonna kill everyone like Skynet is gonna take all our jobs, etc. So let's just put that into perspective, I still believe that AI is a long, long, long way away from where people think it is. So people that have just woken up to AI in the media might think I was just around the corner because those kind of stories sell
Daniel Franco:Elon Musk is your most famous, he's saying at some point, there's going to be existential threat.
Matthew Michalewicz:Yeah. But then there are people like c that talks about the singularity and so forth. But let me give you a counter kind of point to that I spoke at an event where Ray Kurzweil spoke about the singularity, and how it was imminent exponential growth of rates of innovation. At the end, you
Daniel Franco:Could you just explained singularity?
Matthew Michalewicz:it's the moment really where, in the context of AI, where computational capability or artificial intelligence that is artificially created, meets that of, of humanity, and then exponentially grows from there. So the example that he uses of is of the village idiot, and in natural evolution, the village idiot might take, you know, hundreds of generations to improve an intelligence if it ever happens, but AI can start off as a village idiot at 9am on breakfast, and then the Einstein by lunchtime, so to speak. So those are the examples that I use. But I presented a counterpoint to all of that, which is the following. First of all, a personal view. I think scientists in the world that study AI fit into three categories. The third first category is Elon Musk Ray Kurzweil, people that believe it is inevitable that AI will reach human intelligence succeeded. It's an existential threat. And so on. Second end, Ray Kurzweil talks about this being 1020 years away, which as a side note, it has always been 20 years away since 1952. Right? So for the very, very beginning, it was always 20 years away, and it's supposed to be flying cars. Absolutely. The second group of scientists believe that it will happen, but it's definitely not 20 years away, it's 50, or 100. But it will happen. And then there's the third set of scientists that believe it will never happen. And it's not for technological reasons. The really interesting thing about the world and you and me, Daniel, Gabriel in the room and so forth, is that we don't know from a physics point of view what we're made of. So we going back more than 100 years ago, we're actually going back 2000 years ago, the Greeks proposed that matter is made of something called atoms, which the Greek root word is indivisible, nothing can be smaller than an atom. That was a theory for a long, long time. And then all of a sudden, we found that there were subatomic particles, there was protons and neutrons and electrons, everything had to be rewritten. But it was all good because all matter was made from these things. And then all of a sudden, they found that there was actually something below that called quarks. And you had all of a sudden quantum mechanics come on the scene, and you had the subatomic particles up quarks and down quarks and so forth. The Hadron Collider in CERN was created, and they built a bigger one. And then they discovered they they built the standard model of physics, and they said, there's nothing more. And God, they could never reconcile quantum mechanics in general, relatively making physicists believe there is something more and it's at lower levels. And hence, string theory came to the table. So the point is that we don't know from a nature point of view, even what this table is ultimately meant. We're still we're still figuring it out. Some of the world's leading physicists says that nature has created the universe in such a way that it can never be discovered. And the example that they use is to get down to what they call the Planck scale, the lowest level of matter, you would need to build supercollider, that's as big as the universe. And when you switch it on or create a black hole that swallows the whole year for us, right, so pretty amazing statement that nature is constructed in such a way that it's unknowable. So now let me transfer that point to the brain, the brain is ultimately made also from that which we don't know how it works, we have understood only the first layer of the brain, which is neurons synapses, and we haven't even understood it perfectly. neuroscientists and micro biologists are rethinking and reformulating basic concepts on memory formation and recall things that they we thought, you know, 30 or 40 years ago, were understood. Say, we model that first level of the brain and we understand it perfectly, which we don't, the only thing that will happen is we get to the next level of the brain, which they already know, there's these microtubules. And there's trillions of them. And yes, it impacts thought processes, but they're not sure why. Once they've mastered that, then you have the next layer until you reach the same problem you've reached in physics, where you don't understand really what the physical thing is made of. But yet you're trying to artificially recreated at the same time. So I'm a huge believer in AI is not going to reach any level of hype, like the realization of the hype or of promise into we actually made brain breakthroughs in brain understanding and brain function. So I always tell people go and look at state of the art in, in neurological understanding and research around actual how the brain is structured and how it functions because that will need to be understood before it can be artificially recreated. And hence, I think people have really jumped the gun and they've said, Oh, it doesn't matter how it really works. It doesn't matter. We don't know what physical matter is made of. It doesn't matter. We've only mapped the first level of or even mapped and distorted the first level of the brain that none of that matters. Somehow we'll create a master algorithm that is just as good as the natural biological and that's a huge stretch and jump that I'm very pessimistic on you know, it happening the way
Daniel Franco:I look. People have no Look, I must admit, when it comes to AI, I'm scratching the surface level you obviously have a much deeper understanding and the reason why we're sitting here talking about the the thought process, though, that most people go down, or I believe that most people go down is It's almost like with AI. If we say we want this machine to do something in a similar way, you know, it's like the old Aladdin and the Genie, the three wishes type scenario. Yep. Mr. machine, can you please make me the richest person in the world? Right? Or Mr. Genie? Sorry, can you make me the richest person in the world and all of a sudden they kill all the human beings in the world, you are now the richest. Right? So it does something because it's not able to understand the concept of, of humility, empathy, compassion, understand what the values of human humans are. And so therefore, it just, it's still reaching the result that you asked for, but not in a particular sequence that you would have actually preferred? So is that something where you believe most people fall down it just then only really scratching the surface level? What AI actually is?
Matthew Michalewicz:Yeah, my view is that a, let me again, take one step back, they've separated AI into these two broad buckets. They call one general AI, which is the original promise of AI, it's a machine that that can think is aware, can reason and shows every element of, you know, human cognition. And that, in my view, is massive science fiction at this point. The second area is narrow AI, or specific AI, which is the application of AI algorithms like deep learning to very specific business problems, like Siri on your phone, for example, or setting the prices and promotions and celebrations and sip and save. So if we, if we look at it that way, narrow AI is making lots of progress. And it's actually generating great results that our whole business is based on that. But in the end, those are very stupid systems were compared to what general AI is supposed to be doing even the best state of the art algorithms from Facebook and Google around image recognition, I read an article that they still confuse people for gorillas, school buses, for ostrich, mistakes that no human observer would would make. And those algorithms are so highly tuned to recognizing this or interpreting this. They don't carry over to, you know, it's not like general intelligence. So the going back to your first question about the existential threat, there are so many algorithmic challenges within these fields of AI, I had a call from a person yesterday, who was trying to commercialize the technology. And they've been trying to train AI algorithms for three years to recognize handwritten symbols. And the accuracy is at 80%. As a human being, you'd have 99% in terms of recognizing they've tried all these various things and are quite disheartened where it is. So that's like the reality of some of these things. But yet people are talking about existential threat. Yeah, there's, there's a massive disconnect
Daniel Franco:early, but you're also discounting the simple fact that the end, but the I know a lot of engineers, right, people who are creating and designing some really amazing things. And at the core of most of these people, is they want to do good. Yeah. Right. They want to they want to create, and I'm going to keep people safe when they create. Yep, so we are definitely discounting the simple fact that I might end up in the wrong hands, I get that. But most people who are engineers and have that level of knowledge, it is my belief that they are trying to create something that is of value to the human existence, not to take it away.
Matthew Michalewicz:I and I agree with you completely. And I think if we talk about existential threats, I think there are much bigger threats than AI for example, nanobots that they're experimenting with things to put these very microscopic almost particles in oceans to clean up oil slicks, but programmed in the wrong way instead of eating oil, it might begin eating carbon based things into it eats the whole planet. Another example is the from Terminator, Skynet, you've read try to eradicate mankind, you don't need AI to have that kind of scenario, you could actually just have a basic rule based system. And you empower just like a bank empowers a system to block a credit card transaction or to approve a credit card transaction, you could have a rule based system that is empowered to launch attacks or to or to make certain actually didn't have to even be based on AI those I think are the real threats because they're real, they're they're things that are here and now and and not based on speculation that will decipher the secrets of the brain of nature and of physics and then build a master algorithm and then somehow give it control of, you know, military machinery or whatever the case and then there'll be an, there's a lot of you know, if then buts, maybes etc. So there's some real threats those as
Daniel Franco:simple as just turning it off. Okay, can you
Matthew Michalewicz:Another interesting subject that just send the electromagnetic pulse Carrington effects of, of solar flares. Yeah, right wipes out anything technological base. Yeah. And so on. But can you turn it on? It depends how its deployed the thing and terminators that went into every system in every environment. And you would really have to have like an electromagnetic pulse over the whole planet to wipe it out to wipe everything out. It's kind of like a virus of, or cancer in a body. If it's localized to one area, it's easier to deal with than if it spread to everything into your lymph nodes into your blood and, and so on. You're much more difficult then.
Daniel Franco:there's a book written by a lady by the name of Pamela McCall McCorduck in that book. It's, it's called"Machines Who Think" [inaudible audio] was written back in the 70s, I believe. Yep. She said the artificial intelligence began with it with the ancient wish to forge the gods. There's a new book out recently as well. But, Yuval Noah Harari calle"Homo Deus". Homo Deus actua ly means God humans.
Matthew Michalewicz:Is that the same guy that wrote Sapiens, great book, right. Great book. It actually, as Sapiens ends on I think the next step in the evolution of human being a merge of with technology.
Daniel Franco:Well, Homo Deus, his next book after Sapiens, which discusses what his thought processes are, of the humans and of the human race. I guess what I'm asking is, do you are we actually creating something that is greater than us? Is that the next step in human evolution?
Matthew Michalewicz:Yeah, I think, yes. But I wouldn't answer that question. Yes. From an AI perspective. So I think, more interesting area is the merging of technology and biology together. Yeah. And I think this look at technology over the last 40 years, if anything, it's gotten closer to us, you know that you had a computer than the laptop got closer. You've got tablets, you've got phones, everything's, everything's coming close. And I think Elon Musk says that the interface is too slow, you know, typing and so forth. But it would be nothing to do with artificial intelligence, if you could implant a chip into the brain and have that chip interact with the neural link in it, which is Yeah, but but but to enable telepathy in the sense access to the internet, to do away with the device to have to have direct access. It was like the matrix where you plug in and download the content you want, and talking about philosophical discussions, where that could go in the future is that the whole simulation theory that you know, once you connect biology and technology, the most popular video game in the world will be one where you sit in a chair, they put you to sleep, and you're actually born on some planet, and you have a whole experience and then you die, you wake up, you know, a minute has passed, and like Total Recall, yeah, that kind of thing. So I think this area is phenomenally interesting for a couple of reasons. One, lots of research already exists. I think even going back 10 or 20 years, there was studies by was at MIT or one of the major universities, where they had three groups of rats, the first group ran through a maze and had an average time second group had their short term memory cut out from their very cruel experiment. But they were almost lobotomized from a memory perspective, and they couldn't complete the maze, they got lost in the maze. And the third group had the same medical procedure as the second group, but had a special, let's call it random access memory chip designed specifically for a rat brain, grafted on somehow, and they outperformed the first group. And these are real experiments that have already taken place in the past. There's great Scientific American articles about the military, taking control of biological like insects, because for surveillance reasons, if you make a small camera and flying device is very limited by its weight and battery life, but not a biological insect, and they talk about all the experiments and how that's gone to human beings. And and then if you park that to the side, all the studies that are being done around people that have lost limbs, to have an artificial limb and being able to control it, and so forth. So where all of that will head in a very dangerous way is the augmentation of the human body for performance reasons. First, don't happen for military and for medical reasons. Yeah, but then imagine being a stock trade or a lawyer or even a writer and being able to have a medical procedure where you have photos of you know, photographic memory. forever. You can access any piece of information, any case law, your thesaurus,
Daniel Franco:Black Mirror, have you seen?
Matthew Michalewicz:Like, yeah, yeah. And and going back to the driverless car example, at the very beginning, that will be illegal in most countries, but not all countries. And you'll be able to go to one of those countries and have an experimental procedure that you could probably die, it would be a large likelihood of dying. But for some, that's going to be worth it as well. And then over time, get the round to
Daniel Franco:be one. Second release or third. so be an early adopter in that world.
Matthew Michalewicz:So this is so this is why when I hear about AI being a threat to humanity, that might be true in the far future. From now, I can see a lot of more immediate threats that I think you're you're losing our humanity through this blurring of technology, and biology, those are more concerning things than, you know, Siri doesn't work properly on my phone. And I want to smash it every time I do something. This is from a trillion dollar company. Right? And in the same breath, we talk about existential.
Daniel Franco:It's, it's the creative human mind always going to worst case scenario, isn't it? Yeah. The interesting pace for me is, is when we were talking about talking about the next step of evolution of human beings, and based on what you said about the human body, and the upgrades, I guess, that you could make to our body is are we is AI and even just technology, I guess in its own right, proving that we're merely just a vessel, our bodies are just a vessel and then we can upgrade is that something that people are thinking of or, and when I talk about the evolution of human beings, into, you know, we could potentially upload our consciousness into into AI or epsilon machine. It the next level of are we the next gorillas, you spoke about gorillas, are we the next gorillas, gorillas gave birth to Homo sapiens, essentially, there's a lot more science behind it. But in the most basic form, gave birth to Homo sapiens, we then evolved as a smarter being hasn't really worked out well, for the gorillas, they're now second in charge, I guess we've taken over the world. Is that where we're heading as well, in from
Matthew Michalewicz:Yeah, who knows where we're heading. But you've touched upon another very difficult unsolved problem, which is called the mind brain problem. And the question really is, is your you your consciousness, your experience, etc, purely a biological phenomenon from the how the matter is constructed? In your head, the neurons, snap structures, etc? Or is there something you know, in religious circles, they call it the spirit or the soul, etc? But is there something beyond the actual physical brain? Right? Is it the same? Or? That's an unsolved question, and, and one that is, I think scientific America called one of the 10 hardest questions of all time, or hard problems of all time. So if they're the same thing just figured, you know, for argument's sake, then it should be possible at some point to be able to connect technology and biology and just have a transfer from liquid wetware to hardware from biological brain to and then, you know, in the future, maybe even clone your body and download it back. You could live forever. Yeah, in that kind of environment.
Daniel Franco:There's a really good TV, I can't remember the name of it. There's a really good TV show, which is it their bodies are just a vessel. Yeah, they created and only the rich have access to these to these vessels, I guess where they, their bodies are just recreated and when they've broken their leg, they will just get a new one off the shelf and plug ourselves in. And he said that island, there's a few moves on the island is good.
Matthew Michalewicz:replacement. Yeah.
Daniel Franco:It's a new one. I can't remember if it's on Netflix, I know that much. But it's it's quite good.
Matthew Michalewicz:Yeah, no, they're all it shows like that in movies like that really expand our thinking and perspective and
Daniel Franco:what makes it so like it gives us go set a goal to achieve in the inner sense, doesn't it because it's the creative mind.
Matthew Michalewicz:And then well, it also from I'm a fan of ancient civilizations, and I've traveled to all the famous locations like the Nazca Lines and Easter Island and so on. But the interesting thing about the Egyptians is that they believed that if you lost your body, you would be lost forever, which when I was in school, they laughed at them and look at this, you know, superstitious, ancient people that knew nothing, but actually they knew something that we didn't know at the time. If you lose your DNA, you'll be lost forever. So so it actually some of these things cast recast the past on what they might have been thinking, why was it so important to preserve the mummy in the body in a mum mummified way, in the best possible condition that it would last 1000s of years and potentially enter another world, etc. This is beginning to Yeah, it's like Jurassic Park, you find a strain of DNA and so forth. But But on the other hand, if the brain is just, you know, one thing and what makes life in our existence special with something else, something that sits beyond the brain just does, like all the religious people would believe, then there probably be no hope for ever uploading that. Because the thing that makes you you is not actually in your biology, so to speak. But these are such unknowns. And and we're a long way off. Yeah, we're, we're long, long way off. It's, it's kind of like you have to die to find things out. And and maybe when you die, you don't find anything else. Because it's blackness forever.
Daniel Franco:When you did, yeah. There's the notion, we know that we're just merely started out so our body breaks down and goes off into the universe. And we just form into something else later, not not as a piece, but our atoms and matter is then spread across the universe, like the ashes and grab onto something else. And something else
Matthew Michalewicz:Not many people really appreciate why we are all stardust. And that's because in the early years, and by early years, I mean, you know, hundreds of millions of years, there was really only gases, like materials, there was no iron, there was none of the substances from which were made. And you had to have these light materials form stars, the stars had to burn through all their energy, collapse into heavier materials, and finally explode. And when they exploded, they spewed those heavier materials all around the universe. That's what planets were formed from, what what life comes from, etc. So every single human being is comes from a star. It's an it's not a figurative thing. Literally, we come from a star.
Daniel Franco:Yeah, it's fact. It's fact. Yeah. Because what's in the stars?
Matthew Michalewicz:That's right is what is. That's what I remind myself when I go to bed after a tough day. Hey, I'm from a star. Yeah. Yeah, well, my wife say go to bed. Shut up. Stop drinking.
Daniel Franco:Do you do have any major concerns about where we're going? And know you have mentioned sort of, but is there anything that that you feel is one that we should really start concentrating on now?
Matthew Michalewicz:Like, within AI within? Yeah,
Daniel Franco:so what concerns for the world that AI could improve? I should, yeah, clarify that. So environmentally? Could
Matthew Michalewicz:I have lots of concerns about where we're we work on things? going, because capitalism is a system that creates super rich and super poor as we're beginning to observe. And this is why you have radicals being elected into power, because there's just so many people that have much less than their parents had, you know, you look at in the United States, the number the have nots is growing, that people are worse off than they were before. But yet the super rich are exponentially going up. So capitalism creates this divide. And that's a very dangerous and unhealthy long term trend, which which a lot of bad things can happen. In terms of concerns about artificial intelligence, which are things that we should be focusing on. Again, I don't think it's AI. What I'm concerned about it's related kind of similar is about giving systems too much control. They don't have to be AI systems, it can be any kind of system. I had a conversation I think it was today or yesterday with someone and we talked about how 40 years ago people really use their brain to do equations and arithmetic etc. And you know, once you've had calculators become affordable spreadsheets, etc. How many times have you been in a store and someone you know the the kid behind the counter has trouble adding numbers together or subtracting everyday becomes so dependent on technology that basic skills have you wrote it? Same in production plans. 30 years ago, people could create a plan, create a schedule, understand constraints, Theory of Constraints, operations research today, you know, not to down talk people that do ifficult jobs, but they rely on uttons and applications. And hen those buttons and pplications failed, very few of hose people can say okay, hat's all good. I'm going to anually recreate the whole hing. We'll do it old school hat's been lost. So look at riverless cars, people Talk bout in 50 years time people on't know how to drive cars nymore, right? We lose skills s as a calculator or preadsheet or software system, driverless car begins taking ike you don't know how to ride horse, I bet right? What do I now,
Daniel Franco:when I went horseback riding, they gave me the one with a U shaped. Yeah,
Matthew Michalewicz:I haven't even been and I'm sure I'd get the same thing. But 100 years ago, that would be unheard of that we wouldn't know how to ride a horse, right. So we're
Daniel Franco:still comparing the car to the horse power.
Matthew Michalewicz:That's the horsepower. So you look at if we had any kind of major global events, you know, some Meteor hitting or a Carrington effect, etc, solar flare solar flare, the actually people with a basic survival skills that are the most value most are going to live off the land. Correct! Hunt, can filter water, all of that kind of stuff. So the thing that I kind of fear is that you create this level upon level of automation, where it errodes just the basic skills, everyone becomes some ultra specialist in something that in the end, is completely disconnected from what we started off with as a species
Daniel Franco:Talk goes into the useless. What did I say before it was the overuse? No, the useless class? Yes. Where we think we know. A lot. Yeah, because we can push a button. We know how it all goes together. But really, we really lose the knowledge of the basic.
Matthew Michalewicz:I agree with that. And I don't think AI is the perpetrator or the bad guy.
Daniel Franco:It's humans being lazy.
Matthew Michalewicz:And, and capitalism wanting more and more automation. Let's look at factories 100 years ago, look at factories today, look at agriculture, 100 years ago, look at it, automation is widespread. It's not a company I don't go into that doesn't want to automate, how many? How many businesses? Have you met that say, Daniel, we don't want automation, we want to go back to manual ways of working with more people. Have you met one company like that? No, neither have I. Right. So it's not that they were saying, Daniel, we want AI so that we can, you know, automate or stuff like AI might be a technique, but just the general concept of automation, what, whatever kind of way, so we can be more efficient so that we can have less errors, we can have, you know, all these business reasons why you would do it. So I think just, this is why I always come. I don't want to sound like a communist. Right. But like I keep saying your cap, but capitalism is really the driving force behind all of these things there. It's not that business people want AI for the sake of AI, there's always a commercial outcome always.
Daniel Franco:And is that it's a you mentioned that as a potential concern. Yeah. For you. But yet you're Igonna chal enging here that you are creatin a product that is addi
Matthew Michalewicz:I think so yeah, I don't think that I'm part of the solution, in the sense that we create a product that gives more people more work is just the opposite. But we're just like every one of our competitors and every other product out there. So you're going in and selling software that is going to create commercial outcomes. And every implementation we do has business cases that revolve around automation, and uplifts in performance, improve margin, improved volumes, improved revenue, and so on. Everything comes back. No new CEO says, Man, I want to just implement the technology, because I love technology. I've never heard that it's always the business driver behind it. So businesses is this massive force. How do I do more? For less? Correct? Really? Correct? Correct.
Daniel Franco:So if it is a problem, I'm going to dive into this. If it is a problem for you. Well, you believe that it is a future problem. Yep. Aren't you just creating more of that problem?
Matthew Michalewicz:In a way? Yes. But in the same way, that in a system, if you go against the system, you'd starve you? what's what's the abort? What should I do at home, grow vegetables in the backyard and teach my kids that they should? Correct, you know, grow corn in the back and eat that and spend the next 60 years like that? So I've always raised my kids with the view. And we talked about this last time of how to be an entrepreneur, not from a capitalistic perspective, but how to create opportunities for yourself. So you can exist you can survive. This is I always say that entrepreneurs and salespeople in particular, are the modern equivalent of hunters. 10,000 years ago, if you were a hunter, you could feed yourself and your family. Yes, if you didn't know how to hunt you were you were dependent on a hunter or you die, right. So if you knew how to sell and you know how to create economic opportunities, you are like the equivalent of a hunter from outside. Those are the skills that I'm trying and what am today the hunting revolves automation efficiencies, and and so forth. If you want to hunt that's the stuff that's running around the forest.
Daniel Franco:That's where you have to prove your efforts. But what it also does is it provides you opportunity to you raise your profile, you educate more and more people about the potential pitfalls, where you're going to, if you're just growing the corn in the backyard, you're really only serving the four people in your household.
Matthew Michalewicz:True, true. And just like look at companies that create automation, or might create some level of displacement, they themselves, create jobs and create opportunities, and so on. So it's, I think it's an awful kind of mental process to go through to say, Gosh, what can I do to provide for my family, give my kids a good future, and make sure that you know, every single I don't create pollution, and you know, I don't sell anything that God someone might lose their job of or anything like that. It almost be back to your example of the five people on that runaway train wreck? And why do you always be paralyzed from doing anything? It'd be? It'd be really, really yeah.
Daniel Franco:So it's one that, you know, I often think about is, you know, I'm building a business and I'm trying to grow. And, yes, provide a rich lifestyle. When I say rich, fruitful lifestyle, say, for my family that we can experience and travel the world and do you know, experience some of the things that the world has to offer? You can't do that by sitting on your bum. Yeah, yeah, you have to get up and do something. So that that is definitely but it's also am i contributing. You know, we work with some large corporates, who potentially doing things which might not hit my value system, you said, but they are doing some really amazing stuff for the world and taking us forward. But there there's also some things environmentally, that might not be
Matthew Michalewicz:you'd have to investigate every customer you do business with?
Daniel Franco:Well, you do. I guess they're there. Yeah, there's, there's always the two sides to the coin isn't there? Always, they're doing great because they're raising this and they're doing well this thing into the earth or something.
Matthew Michalewicz:You've made me think, though, there's been a number of deployments that we've done that have actually had the opposite effect. And these were for salespeople, because salespeople in the field, selling low value products have really disappeared over the years think of people selling vacuum cleaners door to door by they don't exist anymore. And there's a lot of fear that a lot of salespeople won't exist anymore, people that are selling alcohol products, food products, buckets of paint, because a very low value, kind of ice age, put an ad online and or do it through digital methods and so on. And yet, we've had a number of software deployments where those sales people have been enabled with AI technology that makes them in like a super salesperson in the sense imagine having all the time in the world to analyze every customer, you have says 1000 in your territory every single day to understand where all the problems and opportunities are, who should be talking to what value adding information you shouldn't be doing. You'd be you'd be superman. Right? If you got that thing. Oh, yeah, and one of our applications COP (customer opportunity profiler), uses an internal company, all transactions, call center transcripts, complaints, missed deliveries, goes to social media, what use items and so forth. And in the morning, SMS is an emails to sales rep have the most important things to do through automated analysis that the system has done not through notes that the rep has put in. And then when the rep clicks it, like it might be a thing like one of your customers is running a big event just appeared on social media for that event, they're going to need these three products from you make sure you call them today or visit, you wouldn't know that unless you had been paying attend, you sign me up. And then they throw it in there with the go into the application, they have a list of all the customers, they select the ones that they want to see. And then the system sequences, the visits when they should see them to maximize that the salesperson that was not a poor performer, but was just human, all of a sudden has a huge step up through this automation of analysis that would have been impossible to do manually. So who's got the time to do that no one's got the we've reached the point where no one can analyze all the available data information to make decisions. So you know, when you've raised that there have been maybe half of our projects would fit into that category where there hasn't been an automation aspect and it's been to upgrade the human beings so that they can actually sell more retain customers have better conversations with prospects and customers and yeah, and so on. And then the other half of projects would have an automation element to them and they would sit mostly, you know in supply chain areas and and so on. But I think there's projects where you actually augment someone's job. And they're able to do it so much better as a huge, you know, good goodness attach absolutely makes you feel good inside?
Daniel Franco:Well, we offer a service that that changes people's lives, right, we work in the world of culture, leadership change, yeah, that world where people within businesses will feel more valued, because there's been an upgrade, I guess, in the way of leadership thinking, the improvement in the culture of the business, if there was a technology like that, that could reach that I could use to reach more clients to ultimately make their people feel more valued, they have a far greater reach, then, you know, me sitting in front of LinkedIn, clicking on it, I could speak to that guy, or girl so that there's the upside for that is that what you're doing is you're actually creating far more connection pieces beyond just that salesperson, yet the person on the other end is, or the 1000s of people on the other end, are getting the benefits as well.
Matthew Michalewicz:Without a doubt, I always believe that in sales, good salespeople, educate and add value help to customers and help them succeed. And to do that, unfortunately, requires some form of analysis. Like you can't wake up and go and talk to six customers, and just spontaneously come up with you know, educational information to give them how to help them succeed, be an expert, it requires good sales requires analysis. And hence the you're not automating anyone's job, because the analysis that should be happening isn't happening. So you automate the analysis. No, no automation happens from a human perspective. But that automation of analysis that wasn't done now makes those salespeople much more effective than they were before. So I think that there are areas where automation is applied to tasks that aren't done by humans. And by applying automation to them, it makes the human on the other end that much more productive, rather than applying automation to the human themselves.
Daniel Franco:Yeah, absolutely. Can I get on to that program? Yeah.
Matthew Michalewicz:That was my goal. To sell you a piece of software, I've been sending you subliminal messages. As I walk in the product was on the wall.
Daniel Franco:The piece in all that, where we're offering value, and we're offering data that can improve the human's life I guess we're providing in that same book, Homo Deus by, Yuval Noah Harari, he wrote Sapi ns. Yep. He mentioned that data sm will become the new religi n, dataism, he's calling it data sm will become the new religion where data becomes the most pow rful thing in the world. D you agree or disagree?
Matthew Michalewicz:I don't know. To be honest. Again, when I think of people, I don't think of you and me, I think of all the years I think this 197 countries or 6000 languages being used on the planet, there's even in the United States, you have the whole bell shaped curve from the stupidest people ever created, to the smartest people ever created. So it's very difficult to kind of say, across all of those cultures, all across those value systems and so forth. This one thing will become the most valuable look at how powerful religion is and people's belief in in something greater than themselves and in life, and that suffering in life, rewards itself after death and so forth. And, and, and, you know, for those billions of people to trade that as the most important thing for data ism, Oh, I can't really well, I think, I think,
Daniel Franco:well, when I guess where he's coming from Western countries, Western countries, will send the people will rely on the data for there to make choices.
Matthew Michalewicz:But we've always done that. That's like the only thing that will be different as more data. And we'll just like, it was when we moved from pen and paper to a calculator that was a step up and made things easier when we moved from a calculator or a spreadsheet, that when you're moving to AI to some form of algorithm, which will be like the next computational tool or, or or method because you have to use something more sophisticated and more data is available, and it's more dynamic, but I think all of us have been trying to make decisions based on data. You know, in my 45 years on this planet, I seldom met people it's I don't I don't ever look at data when a decision I don't want to know what happened. I just go totally on my gut and totally I like that. Again, it's not a not a person that you would you meet all of us would would think that we would make you do some analysis. Yeah,
Daniel Franco:that's a good point though. The gut feel is Michelle and I have often talked about this, we believe our gut feel, is generally the right decision to go with, we can do all the the data analysis and background checks. But in some certain certain circumstances where, you know, the livelihoods of people and the welfare and benefits of the people in our team, our gut feel for what's best, the right decision,
Matthew Michalewicz:there's a good book, Thinking Fast and thinking
Daniel Franco:slow, by Daniel Kahneman,
Matthew Michalewicz:yeah, that talks about that. So not to discount, keep in mind that your intuition and gut feel has kept you I mean, it in a collective sense has kept us alive as a species for 1000s and 1000s, and 1000s of years. So when there wasn't much data other than a bush moving in the wind or a smell, or a sound in the night, etc, racking of a tree. Yeah, absolutely. So the people, probably that could make the best decisions, their genetics, are part of our genetics today, people that made very poor decisions aren't here anymore. And their genetics aren't here anymore. But I'll have to think ponder upon that we're definitely in a modern era, where systems are given more autonomy systems are running on data, business people, and people in general are trying to make data driven decisions, evidence based decisions, and, and so on. And that's here to stay, I don't think it's going to go the other way.
Daniel Franco:So going back to your book that you've written, you said, You've co written
Matthew Michalewicz:with my father and a third person, Leonardo Arantes
Daniel Franco:is, is the book structured in sort of from three different angles. You mentioned before that you and your father think differently? Is there? Give us a little bit of background on
Matthew Michalewicz:Sure. attached to this podcast, we'll have the first part of the book, and if every listener can download it and read it, but it's structured into four parts. Part one is really an introduction into AI with chapter one being, what is it? Just like we've talked about what is what is machine learning fit in? is it part of AI where the algorithms said, neural networks, vision systems, and so forth. It paints the picture and tells the story of how the field came to be and where it is today? And what are the challenges and how to look at it. The next chapter applies AI to decision making, which is what our business is all about, and what my father has specialized in for the last 40 years, better decisions, create business, better business outcomes, more profit, more margin, increased market share, and so on. The third chapter takes a specific business problem which we use promotional planning and pricing because of its complexity, and breaks that problem down into what is the data, what is the information layout, what is the knowledge layer, then you move into AI prediction, optimization, and so on. So it's a nice kind of first part that takes a reader from a raw introduction into a very detailed complex business problem written in an easy to understand way, and how AI applies. All of Part two is a discussion on data and algorithms. So if you ever wanted to know what deep learning was, or a neural network, or random forest or ant systems, they're all in there data issues, how to deal with problems that have no data or dirty data or messy messy data, and learning which is a key element of AI. Then the third part is application areas and case studies. And then the fourth part is conclusions and how to bring AI into a business setting talks about things like digitalization, for example, many companies try to jump into AI without first having digitalized business processes and workflows, and really optimizing the way they work and what tools they could be doing it with before applying an advanced level of technology to that so so it takes the reader from an introduction to the subject to an in depth exploration of the details. AI is made of algorithms to application areas written for business people all the way to conclusions, and how do I bring it into my own environment?
Daniel Franco:So your target market is the business reader reader,
Matthew Michalewicz:the CEO, the CIO, or anyone with anyone I mean, it can even be someone doing the job. It can be a production planner scheduler could be entrepreneurs, yeah, it's written for if you have a degree in computer science, then there are probably elements of that book that you would find interesting, which are the business elements that you probably wouldn't be across coming from an academic background. But that isn't jet you would be very across the techniques and how they work and the data and so on. Probably the application areas will be interesting because it gives you the business context. But the Business read and this is my personal observation. AI is hyped, and very few people are everyone talks about it, but very few people really understand what they're talking about. You know, I've been in, in big meetings in really big companies, where they talk about some business problem. And then they say, I will just throw machine learning at that. Next problem, right? Or Oh, yeah, we're throw AI at that. And you and you think they have no idea at all, what they're talking and it's written for those kind of people, not to talk down to them, but just to explain it in very simple language that that when they have that conversation, it's now structured in their head. Yeah.
Daniel Franco:Well, they're making false claims by being able to throw machine learning into it's not gonna work. Is it?
Matthew Michalewicz:what's even worse than not even short machine learning? Yeah. But they've heard it being thrown at things.
Daniel Franco:When it's such a, it's such a deep topic. He goes in so many areas, it goes philosophical goes technical, it goes evolutionary like it's really a such an amazing topic. We are, I'm conscious of your time, we are heading coming to the end. So I don't want to take too much for no problem. Can you name, what's the name of the book,
Matthew Michalewicz:The Rise of Artificial Intelligence
Daniel Franco:the rise of artificial intelligence. And that's to be released when January, January is going to be very excited about this is your fourth or fifth video.
Matthew Michalewicz:I'm not, it's always nice to create a book. And it's a subject that's very close to my heart. But like many things, that when you do something for the first time, like you create your first business, or you write your first book, or what the excitement then is incredibly intense. And then you know, once you've created five businesses or written five books, it's, it's not quite as exciting as Yes, this is more work than you remember. Yeah, that's, it requires more why to process.
Daniel Franco:You mention that before? Yeah, before our podcasts, you mentioned that, yeah. But if everyone was flowing quite nicely, I'm
Matthew Michalewicz:very proud of the material. And that's one thing that you want to do as an author, everything that you put your name on. Same as an entrepreneur, you want to be proud of it, you want it to, to represent something, some standard that you hold yourself to account or some representation of yourself, in your mind. It reminds me of a very quick story from the founder of Lego. And this book might have been brick by brick, I can't remember that I read about the Lego story and how they almost went bankrupt and reinvent themselves and so on. But the story is that the founder, they started off as making like toys and stuff like that. And the founder had a son, his son was in the business, he became a CEO of Lego later. And he found a way to save money, the son by not painting the back of something that was placed against the wall. I can't remember what it was like, maybe they sort of making furniture or at some point, and he went to his father and said, Look, look at this great, you know, savings that are that are made or the paint we're gonna save. And the father was furious, absolutely furious, and said, you know, go back and make sure all of those backs are painted. And and the son said, but no one's no one will see it. No one will know. And the father said, I'll know. And that became that story became kind of like the ethos of Lego of quality. Yeah, how it didn't matter if no one knew or said the quality would never be compromised.
Daniel Franco:Steve Jobs, his same story. His father was[inaudible audio], we step up his and he was adopted. So he's with Father, he claims he was a cabinet maker.
Matthew Michalewicz:I didn't know that when
Daniel Franco:he in through the Steve Jobs, order biography written by Walter Isaacson. Steve was always saying that the back of the desk or the back of the the bookshelf was always the most important because it doesn't matter if no one's going to ever see it. The fact is that it's built properly. Yeah. And so and so when Steve Jobs came out and bought out his Apple Mac, you could see through it. Remember how Yes, yes, yeah. And he wanted to everyone to see that it was built properly. When he came back as CEO when he came,
Matthew Michalewicz:I think it was the iMac. Yeah.
Daniel Franco:So same point, it's it. Do you apply the same for you?
Matthew Michalewicz:I think everyone has, in their mind, even subconsciously some standard? Yeah. For some people. It's a conscious standard, like for Steve Jobs where you know, the founder of Lego, we could verbalize and you could talk about it. For others. It's an unspoken standard, they might do something sloppy and it's okay. And then that's, that's a representation of their unspoken standard. So, so when you said, you know, around the book, it's a lot of work, because for me, I want it to be at a certain level. If I did here, then then wouldn't be so much worse at that point. And really, yeah, correct.
Daniel Franco:Brilliant. Are you going to go on a road show with it?
Matthew Michalewicz:I'm trying to see how long I can go without being on an airplane. Seven months now,
Daniel Franco:before you started, what was the How? You How often did you travel?
Matthew Michalewicz:Every week I traveled at least twice. Once I made four flights that there every single week
Daniel Franco:every single week for seven months.
Matthew Michalewicz:And it's been like that for years and years and like probably since I came to Australia, long long
Daniel Franco:You traveled just around Australia for the
Matthew Michalewicz:very few overseas strip. Yeah, 95%, Australia and New Zealand,
Daniel Franco:because you've got business in all parts of
Matthew Michalewicz:when you live in Adelaide. It's not the epicenter of big business in Australia. I think it's the epicenter of wine and a lot of other great things but but not of big business. So I think you've really got to as an entrepreneur, physically spend some time in Sydney and Melbourne, establish a presence, build a customer base to create a business that has some of those, you know, blue chip logos, yeah, that everyone wants.
Daniel Franco:So you're spending much more time with your family now, which must be great.
Matthew Michalewicz:I've seen what they've seen more of me than they thought they'd ever see. Yeah, they're probably sick of Yeah, I think so.
Daniel Franco:there's not a chance they'd be sick of you mate. Thank you very much for coming on. Again. today. On our we mentioned last time that we might even get a third podcast you will be talking about sales. Yeah. Interesting. entrepreneurs and people with you know, growing their businesses and some areas that they could look at, especially with your now that I understand that you're helping salesmen out in the data world as well, which is yes, which is fantastic. We great to touch on that. We Yeah, we're very thankful for you coming on again, positive that the listeners will love every word as we have. Thank you very much for coming in
Matthew Michalewicz:Pleasure Dan and thanks for having me.
Daniel Franco:Cheers. Bye bye.
Synergy IQ:Thank you once again for joining us here at creating synergy. It's been great spending this time with you. Please jump on to the Synergy IQ Facebook and LinkedIn page where the discussion continues after the show. join our mailing list so you'll know what's happening next at Synergy IQ comm.au. And of course, don't forget to subscribe to this podcast. And if you really enjoyed it, please share it with your friends.