Martin Lucas Podcast Transcript
Martin Lucas joins host Brian Thomas on The Digital Executive Podcast.
Brian Thomas: Welcome to Coruzant Technologies. Home of The Digital Executive Podcast.
Do you work in emerging tech, working on something innovative? Maybe an entrepreneur? Apply to be a guest at www.coruzant.com/brand.
Welcome to The Digital Executive. Today’s guest is Martin Lucas. Martin Lucas is the inventor of deterministic AI and decision science proven across more than a hundred global brands with results up to 76% above market performance.
He has spent 10 and a half years designing and proving a new kind of intelligence when that works with the laws of physics instead of probabilities. His work redefines how machines make decisions and how humans understand their own. Martin’s career spans multiple businesses, five books and projects that bridge technology, psychology, and philosophy.
His research has influenced government thinking, including a paper written for the Prime Minister’s office, and a personal accommodation from the homeless secretary for innovation in behavioral design. Well, good afternoon, Martin. Welcome to the show.
Martin Lucas: Thank you, Brian. It’s great to be on.
Brian Thomas: Absolutely, my friend.
I appreciate it. And you’re hailing out of the London area in England, the uk. I’m in Kansas City, so I appreciate you making the time today traversing these time zones. And Martin, jumping into your first question. You describe your work as going beyond probabilistic AI to what you call deterministic ai and introducing a framework known as decision physics.
Could you walk us through what you mean by deterministic intelligence? How it differs? Differs from current models? What drove you to develop it and why you believe it’s necessary for the next generation of machine led decision making?
Martin Lucas: Awesome. Great question. So where I started in 2015, so I’ve been at this for 10 and a half years, was, why don’t humans understand humans?
Right? And that accelerated through to like, how does decision making work within the brain, how, what’s all the variables that go into it? And that, I’ll, I’ll come back to that and as we develop that forward, right? But deterministic AI solves the biggest problem that sits inside AI today. It’s known in the AI community as the 30% problem.
So every prompt that a human sends, every software that’s run by code has this 30% issue. And what it means is that because it’s a stochastic model and a probabilistic math model, it is inconsistent with what it does. So imagine, for example, you’re a finance company and you want to use AI for loan applications, it will reject some people and it will then accept other people that are exactly the same.
In Pharma meds research, it means that you will get unreliable results. So it could say that this medication is good. One time that you run it, you could run it again, and it says it is bad. That’s the 30% problem. So in market and personalization, you’ve got people looking at customer actions and it becomes all inconsistent.
So what we’ve done with deterministic ai. Is created AI that allows you to be accurate, so it takes away the probability so you get continuity and consistency of results. So it’s not like a choice of deterministic AI or AI as exists. This actually overlays onto existing AI and makes it accurate. So it solves a pretty big problem for reliability, first of all.
Brian Thomas: Awesome. Thank you for sharing that. And yeah, over the last 10 years, we were talking earlier you’ve just done a lot of work in this area, a lot of research understanding, as you said, why humans don’t understand humans.
Martin Lucas: Mm-hmm.
Brian Thomas: Deterministic AI allows for accuracy, reliability, and more consistency.
So I appreciate you breaking that down for us. And Martin, your career spans technology, psychology, philosophy, and you often say that intelligence, human or artificial needs to be understandable, reproducible, and deeply human at its core. How do you reconcile building rigorous deterministic systems with the messy emotional content rich world of human decision making?
How much of your work is engineering and how much is behavioral science?
Martin Lucas: That’s great. So I’m, if it’s okay with you, I’m gonna run a fun test. I believe it’s fun, but you can tell me at the end, right, A fun test for you and for anyone listening to this, right? Because what we’ve done is built intelligence that works with human thought, that understands it, right?
So you’ve got the deterministic AI that makes AI accurate. And then as you mentioned, for the messy stuff, like what goes on inside the brain is what we’ve built with the decision science and decision physics. So if it’s okay with you, we’ll run a little test, right? So is that cool? Brian?
Brian Thomas: Sounds great.
Martin Lucas: So I want you and every, everyone listening to this, just take a breath and I want you to think about a brand that you genuinely love, one that feels right to you.
What’s the first thing that comes to mind when you think of them?
Brian Thomas: Trustworthy.
Martin Lucas: Okay. And when, what do you feel when you see or hear them? Right? You mentioned trust. Is it security? Is excitement, is it belonging? Is it identity? Like, what kind of feelings do you have towards that brand?
Brian Thomas: Just again, trust a relaxed feeling that my experience my next experience is gonna be as great as the last one.
Martin Lucas: Awesome. Thank you. And then what we’re gonna look at next, so this is great. So Brian’s just exploring how he feels about a brand that he likes. So I want you to keep thinking about that for everyone that’s listening to it. Right. Do Brian, when you think about how they, the tone and the words they use towards you, like, do you, do you, are you.
Do you want ’em to be calm, to be bold, to be more human, to be more clever, to be simple. Like what is it about the brand that makes you engage with ’em?
Brian Thomas: They meet my needs. They’re, it seems like they’re a step ahead or every, again, every experience I have with this brand does not let me down.
Martin Lucas: That’s fantastic. It’s really, really fantastic because one of the biggest things about decision science and understand decision making that most brands fail at and the world doesn’t understand is like most people are looking to understand, what problem do you solve for me, right? So as a side note, the average human brain makes 36,000 decisions a day, and only 2% of them are conscious, right?
So you could have a brand that like has all the things that you want, but if it doesn’t speak the way that resonates with you. Then you reject it. Right. So just picking that up again. Right. Do you love everything that they sell or just certain things?
Brian Thomas: Most everything, yeah. And I’m probably would, would look at other things that I haven’t purchased from them, just because they’ve done such a great job with their products.
Lucas-Martin: Cool. Okay. And describe a time when they absolutely nailed the experience, whether it was online, in store, general communication. What made it stand out for you?
Brian Thomas: Just the the experience using their products. Just phenomenal everything from set up configuration to enjoying the entertainment that they provide.
Martin Lucas: I love it. Okay. And when you think about, when you’re engaging with ’em, and I don’t mean purchase, right, because the world is, is very much focused on, on purchase to the cost of like what people want from an experience, right? But when you think about when you want to engage with them, is that like when you’re scrolling at the weekend, is it late at night?
Is it when you’re at work? Is it when you’re on holiday? Like what, what, what does the timing look like for your relationship with that brand?
Brian Thomas: The timing is, or when I most think about it, is probably when it’s time to relax with the family or yeah, just some downtime. And that’s probably when I think about them the most.
Martin Lucas: Nice. Thank you. So we we’re just exploring all these different conditions that go through your brain, right? This is how you’re the architecture of your decision making for this particular branch, right? And it sounds so simple, but if we look at it conversely, right? If they ever done something that made you hesitate, like the tone fell off, the campaign was wrong, they’re just sending you too many messages, too many discounts, like in the instant that kinda broke that magic of your, your relationship with ’em.
Brian Thomas: Not that I can think of, but I’m sure it’s happened, but I just can’t think of an ex, a time or example.
Martin Lucas: All right, cool. Okay, so final, final question for me, and I appreciate you doing this, right? So hopefully it’s been fun to explore that kind of com component, right? When we think about your loyalty towards this brand, and I mean the feeling of loyalty inside the brain, right?
Like is it the product? Is it how they make you feel? Is it when they engage with you? Is it new products, is it new experiences? Like what drives you to think of that brand to want to go back to ’em again and again?
Brian Thomas: They don’t disappoint. They deliver high quality products, and the experience has always been the best.
Martin Lucas: Love it. Thank you. So what we’ve been looking at there, right, is we’ve been examining brand’s, relationship with that brand, his emotional context, his language, his timing, what resonates with him, what experience that he wants. That’s what we’ve done with decision science. So that each brand, and we’ve worked with over a hundred different brands, 120.
Million different tests. We performed 76 above. 76% above the industry, right? Which is all wonderful because what we’re doing is understanding each audience and what experiences they want and what experiences they don’t. That’s what decision science is at its heart.
Brian Thomas: Amazing. And thank you and I appreciate stepping us through those examples.
This will be great for the audience. So Martin, you emphasize augmentation rather than replacement. Designing machines that work with human decision making rather than sublet, rather than supplant it. Right. In your view, how do organizations balance automation and human agency in decision science? And where do you draw the line, especially in domains like health or mission critical infrastructure.
Martin Lucas: Yeah, I think the, I think the text the next 10 years is gonna be very much about alignment. So machine logic and the acceleration of what machines can do and what AI can do is fantastic, but it must understand intent and context. So going back to what I mentioned about deterministic ai. When you’ve got a 30% drift problem, you can’t use that in healthcare.
You can’t use that in finance, you can’t use that in pharma. Deterministic guy reduces that to less than 1% error rate, which is perfect for any kinda governance market. So at the moment, it’s a little bit too wild west and too much of an AI bubble that’s going on with not enough governance and security.
And the funny thing is that governance and security means that you can actually understand humans better. And what meets in the middle for me over the next 10 years is that the rise of human creativity of ideation is gonna manifest in slightly different ways, but you’re never gonna be able to replace that human creative component.
So I think there’s a world of opportunity if we understand how to manage what technology brings while still using the best of what makes humans, humans.
Brian Thomas: Thank you. I appreciate that. Love to get people’s insights here on the podcast, but you talked about the future next 10 years. Is that alignment?
Again, the technology it’s gotta understand intent, it’s gotta understand the context. But what I took away from it though is that rise of human creativity may look a little bit different than it does today, but I absolutely agree. There’s nothing that can replace, a human being and its creativity.
So I appreciate that. And Martin, the last question of the day I have for you, looking ahead even, five, 10 years, how do you imagine the landscape of AI cog cognition and decision making evolving with your deterministic framework? What kind of world or systems do you hope to enable and what legacy do you want your work to leave behind in the way humans and machines interact?
Martin Lucas: Yeah, so I’ll give you an exclusive here. So we, we’ve just signed off the patent and the invention. The most code today works on the basis that you’re, you’re telling a machine software what to do. Right. That’s, that’s traditional computing code. What we’ve built with deterministic ai, combined with decision science, combined with decision physics, is symbolic mathematics.
And what that does is teach a machine how, and once a machine understands how it means it can operate with reasoning and understanding if it’s got reasoning and understanding, it’s not just that it can do things quicker, but it can do things by you explaining and teaching it the context of the people that it’s trying to serve.
So rather than just being able to. To produce lots of stuff, which is lot, what lots of businesses are doing right now, I believe. And what I’m gonna prove is over the next 10 years, it becomes about quality and how you serve people better and how you understand humans. And you can combine that, that mythology of the sci-fi of humans and machines working together.
So I’m excited for what the future looks like. There’s lots of opportunity.
Brian Thomas: Thank you. And you did tease that apart a little bit. Generally, the traditional code is written to instruct AI to perform certain algorithms or certain functions, but your decision science and physics is really about symbolic mathematics, I believe is what you said.
And you’re definitely taking a different approach around this and I really appreciate you sharing that with our audience today. And Martin, it was such a pleasure having you on today, and I look forward to speaking with you real soon.
Martin Lucas: Thank you, Brian. It was awesome.
Brian Thomas: Bye for now.
Martin Lucas Podcast Transcript. Listen to the audio on the guest’s Podcast Page.
