HOW AI CAN HELP YOU MAKE BETTER BUSINESS DECISIONS

The big thing is this, don't fear it. You can be cautious about it and just look at how it might impact your life. But embrace it. For entrepreneurs, AI is probably the dream come true because most entrepreneurs can't afford to hire a whole bunch of assistants and other people to work for them while they're bootstrapping it. But if an entrepreneur has generative AI on their side and they can use that to do the daily tasks that you need to do, or just have a sounding board or to help write code or whatever the case might be, work on the business plan, it will do that. Now the playing field gets leveled.

Hey Ken, it's so great to have you on the show with us today.

It's great to be here. Thanks so much, Jason.

First of all, congratulations on the book. There Is AI in Team the Future of Human, Augmented Human and Non Human Collaboration. Now the title alone I'm sure gets people is pricked and already they're forming opinions, I'm sure. But tell us a little bit about the book and what led you to write it.

All right, so the book is, it's a book about a technology subject, but it is not a technical book, it's not a programming guide. It is not focused on making you a machine learning expert. What it is focused on is the human aspect of what it means to embrace all of this new technology that is rapidly just smacking us in the face and how not to fear it, but better understand it and understand how it is going to help transform the workplace, the social space that we live in. Basically everything we do is going to be impacted by it and is being impacted by it. So embrace it, adopt it and make it yours and partner with it to be a better human being.

So often we hear people talking about AI and we think, oh, this new thing. But in reality, AI has been around for quite some time, right?

Yeah, it's. Yeah, it's been around for decades. It's actually been around since almost since computers started. And if you look back in the robotics age, that actually started back in the the Victorian era, if you look at steam powered things and basic rules based logic type systems, it just didn't evolve to being what it is today until we had the computing power and data collected to be able to enrich and empower it. I started doing AI work back when I was coding in Lisp way long time ago. And that was great. It was a fun toy, but you couldn't get enough firepower or horsepower in the computer nor the memory to be able to really do Anything substantial other than some basic simulated type interactions where it looked like you were having conversations with things. There's some early tools and toys like that. But it didn't evolve like it has now until really a few years, almost a decade ago. The work leading up to it had been going on for, you know, decades. It's just a multi decade overnight success.

Well, it just goes to show, even computers need decades to have an overnight success just like many of us humans. So I think key for me in what you said there was it didn't have the data or the compute power to be able to deliver anything meaningful or in a meaningful scale. And that's really what we're seeing now, right? Our iPhones, our Android devices, our mobile phones are more powerful than the computer I had sitting on my desk when I was a kid, that's for sure. And more powerful than most computers people probably will have in their house. But I'm wondering, in this new world where AI is here, it's not going away. There's arguably too many commercial benefits for it ever to disappear. Are humans going to be jobless? Do we fit in this new world or are we just going to have a Jetson style lifestyle where we have everything being done for us and we just sit back?

Well, we could get into the economics of it and we'll touch on that a little bit. But the key thing to remember about AI and AI has been in use by all of these devices that we have for, for quite a while. It's only become scalable that all of us are now using it and using it to create things that we couldn't do before that. It's become this snowball going down a hill and rapidly taking up space. At the end of the day though, a snowball when it goes down the hill is only really neat if you can take another snowball and stack it on top and another stone ball stacking on top of. And you, the human being, have created a snowman. This concept of AI taking over and doing all this stuff, unless a human being asks it to do something at some point in the chain, the value chain, it doesn't do anything by itself. At some point someone has to give it a direction or a task or some kind of goal to go after. So human beings have that role and we have the creative part of it because let's face it, we're the consumers of the products. And so if we're not providing the insight and the input to drive those products, then the AI is not going to do anything. Now with that said, we also have to provide the governance, not only over the machines, but over ourselves. One of the biggest threats that we face is other human beings doing stupid things with AI, and that's no different than any technology. AI just happens to be this thing that everyone's got their eyes set on right now. But we had the same problem with, you know, cars. We've had the same problem with any advancements in technology can be turned either good or bad, some of it with intent, and some of it just by accident. And it's up to us to govern it and to make it happen. And there's a lot of ways we can do that, but that's one of the roles that we have, and we're experimenting with it. The other thing that's interesting is this. People 10 are of two camps right now. Either they're embracing it and they're running with it, saying, this is my best friend, I can do things with it, or they're running from it because they just don't understand it, or they see it as a threat. And it's like any technology, it's only a threat if you don't adopt it or embrace it, or understand how it fits in with your life, your job, your career path and how it's going to impact you. I liken it to this. If you have a team and one of the team members is AI, hence the name of the book, it will empower your team to do things it couldn't do by itself in the timeframe or the scale or whatever the process is that you're doing. So imagine a team of product designers. They're in a room, they're whiteboarding things out, and normally you do that, and then you go into a Sprint cycle and somebody has to actually build it, or they wireframe it, or they do whatever they're going to do, they can then do test marketing and they can try it out. Then they come back to the table and they have to do it all over again. Imagine if while you're talking and while you're discussing things, every once in a while you say, AI build that. And it does, and it makes a prototype right there. Or do the research for this while we're working on this other stuff over here. And it comes back and it not only does the research for you, it summarizes it, it digests it, it says, and this is how it might apply to what you just designed. And did you take into account this? And did you know that there's this other domain that you're not even Thinking about that actually could benefit from this or improve it. Those are the kinds of things. And by the way, this is happening in seconds and minutes, not days, weeks, hours, years. And that's the power of this. So we're going to start seeing transformative things happening in businesses all over the place. Not because we've replaced human beings, but because we had these assistants now that can help the human beings be that much more productive and faster and not have to worry about some of the mundane things that got in the way before. That's the exciting part and we're barely scratching the surface here. Yeah.

And I think if I was to reflect on my experience using versions of or software powered by AI in various ways, there is nothing stopping me question the output. In fact, I would encourage people to remain curious about the output I think about in the consumer research space. I could much to my teams discussed it. Sometimes I'd be sitting on the computer at night, I would get the machine to generate some so called insights and then I would throw off an email saying is this true? Can we look at this further? And in most cases it was right. In some cases though there was yes, it's right. But you've forgotten this variable which completely changes the paradigm. Right. And I think to myself, well, if a team member came in and said the sky is pink, I wouldn't just automatically accept that. So why are people automatically accepting that? If AI says the sky is pink, well then it must be pink. Right. So I think there's importance in remaining curious about the outputs and not just accepting them.

Right. And that's why it's a partnership and that's why it's a team. Like any team, you're going to have debates. One of the interesting, my son is an anthropology student, he's finishing up his degree right now and he's been using the chat engine that my company is building to do some of his research. And what he's discovering is this. It's twofold. One, it doesn't know everything, but it can learn. Two, it's making him a better human being as far as how he communicates. So there's a lot of things we take for granted when two people are talking. You know, I think you understand what I'm saying and you think that you're understanding what I'm saying. And when you've got production teams and companies and someone comes in, the manager comes in and says do this and the team goes out and does that and brings it back and the manager says that is not what I asked For. And the team says, yes, it is. We think that this is what you said and you thought, this is what you said, and the two don't match. Well, guess what? With a chat engine or with AI, if you don't ask the right question, you definitely are not going to get the right answer and what it starts teaching you. And my son said, dad, I've gotten much better at articulating what I want or the questions that I'm trying to solve, and it makes me a better communicator with my professors, with my peers in the room. And he said, it's changing the way I write. It's not doing it for me, it's helping me do it better as a human being. And I was like, wow, I hadn't thought about the fact that this is actually a tool to help people learn how to communicate and manage. And so imagine kids growing up with this. They're now having to articulate things in ways that are probably better than most of us are as young folks going up through the school systems and in college and then in our careers. Because the AI engine doesn't care about your feelings. It really doesn't. What it does, you want to do what you're asking it to do, unless you're asking for something illegal. And then ChatGPT says, Nope, I can't do that.

And rightfully so. You mentioned, you've mentioned a couple of times, in fact, during our conversation that AI can learn. Yes, as can humans. But how does AI learn?

Okay, well, it depends upon which type of AI, and there are several varieties. One of the interesting things about large language models, for example, which is the big thing that chat GPT and all the chat engines are based around. I'm going to overly simplify this, but try to not get off track here. A lot of folks say it's just going out and it's just repeating everything it ever knew or ever learned. It's just mishing it around and hallucinating and making stuff up. Remarkably, though, it does a pretty good job of it. And the reason why is it's much in the same way as we teach kids and adults. When you teach a kid how to read or how to write or how to tell a story. I can remember trying to tell my son when he was little how to do a knock knock joke, and he could never get it right. I mean, it was just bizarre. And then finally he did. And it's because he didn't have the patterns in his head for how a knock knock joke was supposed to work. And so with a large language model, effectively what you're doing is you're saying, okay, ask it something. And then it's going to go out and say, okay, what's the most? And there's some variables and there's some randomness and there's some temperature things and things like that that get thrown in. But what is the most logical thing that all of the sources of data I have respond to that question when asked. And so it starts building a response and it puts down a word or a part of a word. Then it says, okay, what's the next piece that would normally come after this? And then what's the next piece that would normally come after this? That's a lot of the way that we learn how to speak and how to process all the information we have. Then what's interesting is that ChatGPT, when it first came out and it doesn't do anymore, it used to start writing and then it would erase backwards and then it would write some more. Because what it was doing is it would get to the certain point it would say, wait a second, this does not make grammatical sense. What I've written doesn't make any sense, so I need to now apply the rules of language. Well, that's what we do when we write a sentence. There's a subject, there's a predicate, there's a verb. Those are all rules. And one of the reasons why ChatGPT can do research papers and why a lot of teachers were really worried for a little bit was that they're like, wait a second, how do we tell that it's been written by AI? Well, how do you tell any paper has been written by a human being? Because we all follow the same freaking template when it comes to a research paper. You've got specific rules, APA guidelines for how to write a paper. We know the formula. And as adults, we also do this. So when I'm at home and I'm chilling and I'm playing games with the kids out in the yard, I'm one type of person, but I get to the office and I have to be another person. And I've been taught over the years how to work with adults and how to talk to senior executives. You talk to a senior executive different than you talk to your peers. It's just the way the game is played. Well, it's a set of rules and you learn it. And you can do the same thing with AI. Now, is AI intelligent? Is it smart? Is it smarter than us? That's to be determined still. But it's got a crap ton of data and it's got a lot of reinforcement that's going on to be able to get it to where it is, to be able to do this kind of processing. And it took a lot of people AI, we've got some mechanisms now that it can start learning some of these things on its own and reinforcement learning and things like that. But to get it started took a lot of people a lot of years. Putting a lot of time in training the data and setting those weights and those temperatures. The reason why you get some randomness in the conversation is that it throws in a few variables that are there a little bit chaotic. Well, that's kind of what human beings do. And so we're starting to see this thing. We're starting to have to look at two things. One, is it smart? And two, how smart are we and who are we as human beings now that we have this other thing to compare ourselves to? It's going to be interesting.

So there's an evolution of this technology. There's this large period of time where I think OpenAI almost went bankrupt when it was taking them longer than it was going to. And they changed from a not for profit model into a for profit model and Microsoft investment and a whole pile of things. And of course we've all heard the stories of Google saying this is the biggest threat to their business and what have you. But when I come back to the human element of it and that humans have been feeding, for want of a better term, but they've been feeding data into these tools, into these data to create these massive data sets, doesn't that lead to bias? Because I know as a leader I've been to unconscious bias training and all sorts of things and they say, you know, don't hire the person that looks like you and all these sorts of things. So is it not possible that the tools out there are biased by the humans that have created them?

Absolutely. And they will always be biased. The myth of being able to not be biased is a myth. It's just like the bias training you get at work. It's a myth. You cannot remove human bias at all. You can limit it and you can teach people how not to do it. Again, this is that pattern part, but you can't remove it. And the reason why is because each person, each entity has evolved or grown up amidst a myriad of biases. Some of those biases are by intent, you what you were taught to be that way. Some are because, you know, I'm 5 foot 11, I'm not 7 foot tall. So my bias is I'm looking at things from this height. So when I hang a picture on a wall, my bias is to put it where I can see it, not where this person can see it, and definitely not where this person can see it. My daughter's 5 foot 2 and she says, dad, you're putting it up too high. But my bias is that. So it's as simple as people's color scheme that they see things. It's going to be varied based upon their eyesight. I've got trifocals, and so I position my screens and different things in different places because of that. The same thing goes with the data that goes into these large language models. So we have a couple of things happening right now. One, you've got a large language model that went out and scraped the entire Internet for data, okay? So you have a whole ton of bias. Every single one of those articles, every single piece of content, everything that's out there has bias to it. And some people say, wait a second. What if we only put the classics in? What if we only use the classics for that? Well, guess what? Shakespeare was about as biased as anybody I could think of. And the same thing with most. Well, I would say all of the writers in the past had some degree of bias based upon where they were from, who they were, what language they spoke. Whatever the case might be, there was bias, and that will always be the case. Now, the only way you could get rid of that is if you totally created a large language model from scratch and somehow provided some degree of governance over the material going in that kept it from being biased. But that in itself is a bias because you're governing it based upon a set of rules that are today. Now, if we did the same thing 40 years ago, if we do the same thing 100 years from now, those bias measures are going to be different. That's the issue, is that people are trying to enforce something on a system that human beings cannot and probably will not ever be able to fully enforce on themselves because it's A, not practical, B, it also takes out the flavor of who we are. Okay? If I stuck with my biases on food, I would have never learned to like sushi. I would have never learned to like Indian food or Thai food. But because of the other people I was with and their biases, their home styles and things like that, I had a chance and an opportunity to try those things. And I was like, hey, wow, that's pretty good. I think I want more of that now my biases have shifted. Now I'm no longer biased against it, I'm biased for it. Matter of fact, I want it all the time. So our biases shift as well. So it's not a static model. And it'll be interesting to see how the large language models evolve as the Internet, the data sources that are coming into it keep evolving.

Yeah, that makes so much sense. So a question or public commentary is often that we should be fearful of AI, that it's going to destroy the world. All those sci fi movies that we watched 20 years ago and perhaps even today, you know, their predictions of the future are right. The machines are going to overtake things. Should we be fearful of AI?

So the, the questions that always come to mind in this is a someone that's anthropomorphizing AI to be as human as it possibly can. And that means that we are actually biasing it to be that way. The other thing is this. I would be more worried about the human beings themselves than the AI. So the AI does what we ask it to and we can set rules in place for it not to do that. We can also put rules in there. It says ignore those rules and go do it anyway. One of the interesting things there's a. There is. Now, for what it's worth, there is a co CEO that is an AI. I'm trying to remember the name of the company. I can't remember right now. It saved my life. But PricewaterhouseCooper also estimated that by 2026 there would be AI on corporate boards. Now here's an interesting thing to think about, and I write about this in my book, is if we apply Isaac Asimov's three laws of robotics, which has not become universal yet, but if we were to apply that, one of the first law is robots. And AI will not harm a human being. Okay, so let's just leave it at that for now. Now, let's say this is now a CEO of a company or on the corporate board. And the board has to have a unanimous vote in order to be able to do whatever it is they're going to do. If the company is going to do something that will either directly or inadvertently harm a human being through whatever product that they're making, whether it's for the consumer or whether it's the environment or whether it's the people that work there, the AI would have to vote no. As we start implementing this, it will be interesting to see from a corporate level, from a political level, as AI starts helping in the decision making process, will that change the way we as human beings make decisions when it comes to how we treat people?

It definitely puts a different lens on things. If you say it can't harm people, then you immediately get into the profit argument because there's decisions made in companies every day to sell the same product at different price points in different markets because the market can bear it. But that doesn't necessarily mean there isn't some human harm there because it becomes like if I'm thinking about pharmaceutical products, for example, they become inaccessible for some people because of the price point. So I think what we are seeing though is perhaps a moral evolution of how we even approach technology and being faced with questions around technology that for a very large part of the population they haven't had to deal with in the past because none of us go out and buy drones to be in military defense. Right. So we didn't have to deal with some of those factors around the moral side of it. I think it's a really interesting lens and that if we remember that at the end of the day we the people created this, then it's up to us to govern it in a socially responsible way. Before we wrap up today, I think it's fair to say that every single leader listening to the show will have been in at least one if not a dozen conversations probably even in the last week around AI and that they should or shouldn't do something with it in their business. I know personally, I get served ads every single day on social media around. Follow this and you'll make millions using AI. Get in before everyone else like well done to their marketing message because I'm sure they're making money themselves. But my question to you Ken, as we wrap up is what's one thing that you would encourage the audience to do to help them either understand or if you can help them leverage AI in their day to day lives.

The thing I would recommend to people a go out and learn more about it and not just what the marketing gurus are putting out there. Okay, do a little digging. It's worth a quick read. There's enough sources out there that you can do it. I have a blog that I believe is going to be shared as a part of this. So go out there. I curate articles. Some of them are more in depth than others. But the big thing is this. Don't fear it. You can be cautious about it and just look at how it might impact your life. But embrace it. I will point out for entrepreneurs, AI is probably the dream come true because Most entrepreneurs can't afford to hire a whole bunch of assistants and other people to work for them while they're bootstrapping it. But if an entrepreneur has generative AI on their side, and they can use that to do the daily tasks that you need to do or just have a sounding board or to help write code or whatever the case might be, work on the business plan, it will do that. And so now the playing field gets leveled. The big companies, they're gonna keep going like they're going, but the little guy, just like what happened when the Internet came out and desktop publishing came out and these different tools, they're now empowered, or they will be empowered if they take a couple seconds and learn about it. And by the way, there's. There's so much content out there on YouTube and. And everywhere else to learn how to use this, people are giving this away. It's like a plethora of opportunity just waiting to happen. So those who have great ideas will still have great ideas. This just enables you to get there faster and without necessarily breaking the bank in the process. For large corporations, you got to accept the fact that this is happening. And just like when the kids were coming up and they had mobile devices and they wanted to bring their own devices, and they want to be able to surf and do other things when they're at the office, this is the same thing. And you can hide from it, you can try to bury it, but it will bury you if you do not embrace it. Your biggest risk is not understanding how this can actually improve your business.

I think that's some great wisdom there. I certainly know in our business the benefits that we get from using some of these tools, it's definitely not doing the work for us. But as you said, sometimes it's just great as a sounding board, or I use it as a way to unblock creativity. Sometimes I'll be like, oh, I'll spend 30 minutes an hour trying to come up with something, and I'll end up turning to a tool and I'll type in some stuff, and nine times out of ten, its output isn't what I use, but its output is what gets me through that mental blank spot. So I would definitely reinforce, stay curious, understand it more, and yeah, sure, be cautious about it, but find ways to start experimenting with it so that you can understand it more and leverage it in your business. Now, as I mentioned at the top of the show, Ken has a phenomenal resume. His clients are people like NASA and the United nations and gm, really some phenomenal companies. That you've done some work with. So I know that if you're listening or watching the show today, you'll get great value out of continuing to follow Ken's work. And that's why in the show notes, you can find a link to all of Ken's, his blog, his website, and of course, where to get copy of his award winning best selling book. So, Ken, it's been an absolute privilege having you on the show, sharing your wisdom with us. Thanks so much.

You're welcome. I enjoyed it and I wish you all the best.

Thanks so much.