Business Innovators Radio - Episode 34: Building the Future: AI’s Evolutionary Role in Construction with René Morkos

Episode Date: September 30, 2024

Construction Executives Live Series Join us for a captivating live discussion exploring the groundbreaking impact of artificial intelligence (AI) in the construction sector, featuring René Morkos, the founder and CEO of ALICE Technologies, and an esteemed AI expert. Throughout the episode, we’ll tackle pressing questions, illuminating AI’s practical applications in addressing industry challenges, including labor shortages, accelerating tech adoption, and optimizing scheduling processes.In The Zonehttps://businessinnovatorsradio.com/in-the-zone/Source: https://businessinnovatorsradio.com/episode-34-building-the-future-ais-evolutionary-role-in-construction-with-rene-morkos

Welcome to In the Zone and Construction Executives Live, brought to you by U.S. Construction Zone, bringing you strategies for success with construction innovators and change makers, including In The Zone peer-nominated national award winners. Here's your host, Jeremy Owens, Live. I'm your host, Jeremy Owens, owner and founder of U.S. Construction Zone and three generations improvements out in sunny, Northern California, where we will have 80 degrees today. Welcome back. I see a lot of familiar names and faces. Thank you for being here. We have another great show for you today. We're going to get into technology. I know this is one that I get a lot of questions about. A lot of us in construction are struggling with this, whether we're slow to adopt it,

whether we've adopted and gotten burned before. There's a lot coming at us. And I know if you're like me, I like technology. I like to add things to my business to make me more efficient, but sometimes it's hard to choose. So we have someone that's going to help us dive into the subject of technology and especially AI. I can do chat GPT, but that's about my limit right now.

But I'd like to learn more about how I can implement it in my business, talk about it with my members and my network. So we're very fortunate to have Renee on the show today. We are sponsored by the great team at Bill 12, speaking of technology. They are a great marketing machine. CRM, you name it. They make me more efficient at three generations improvements. They really have replaced a human in terms of marketing efforts and automations and really valuing each lead that I get. So it's been a great partnership with Build12. Check them out, Build12.com. Our show today is titled Building the Future AI's evolutionary role in construction.

Our guest was on my podcast two and a half years ago, and so we're welcome to have him back. René Marcos is the founder and CEO of Alice Technology. He obtained his PhD in artificial intelligence applications for construction at Stanford University. He is a second generation civil engineer with over 15 years of construction industry experience that is divided between industry and academia.

This professional experience includes working as a project manager, in Afghanistan, underwater pipeline construction, automation engineering on a $350 million gas refinery expansion project in Abu Dhabi, ERP system implementations, and various virtual design and construction projects. Please help me welcome back. Renee, Marcos. Renee, thank you for being here. Jeremy, great to be back.

Yeah. Two and a half years in real time, something like 10 in startup world. I was just thinking, like, it is. felt like maybe a, maybe six months since we last spoke, but that's kind of where we're at, right? Everything is so rapid and so fast. I mean, what you're dealing with. I don't even know how you keep up with it. Yeah, the law of accelerating returns. So the head futurist at Google was a guy called Ray Kurzweil. And he wrote this manifesto back in 2001, the law of accelerating returns. Ray Kurzweil, look it up. Probably the best 40 pages you'll ever read. He was, he was, he was,

He basically said that like the rate of change is going to continue to improve and increase, like faster and faster and faster. It's going to keep adding up on itself. And so, yeah, it's one of the challenges that anybody in the AI world faces today. We went from, you know, 10 years ago, like, what is AI, is anything happening to now? It's just like it sounds like there's a startup every week. Yeah, no doubt, no doubt. Where are you residing now, Renee?

I live in New York these days. I did my 14 years in the Bay Area. I've been there, done that, got the T-shirt, spent about 14 years watching the valley do its thing through its ups and downs. I've been very privy to like some of my friends that have been on those roller coasters and myself, watching like bleeding edge tech go from like lab to company to prototype to company, right? Yeah, yeah, crazy times, crazy times indeed.

So for those that don't know you, give us a little. a little bit of background about how you got started in this industry? Pretty easy. Dad was in civil engineering. Okay. I graduated high school. Graduated at 16. And my dad gave me a good piece of advice.

He says, son, study anything you want. Just don't do civil engineering. I was like, got it. I was lost now. I'm found. That's exactly what I'm going to do with my life. So I want and studied civil engineering. I'm sort of joking.

I don't think it was like just rebelling against my dad. I liked his job. I liked the fact that you could work globally. I like the fact that you could work outside. Yeah. I like, like, I like building things. I really like the concrete and the hair. I find something like enamoring, something like magical about you go home.

And I've had this like several times. It's just this weird thing. Like, I'll sit there and I'm like, there was no concrete slab in the morning, and now there's a concrete slab. And that concrete slab is going to be there after I'm not. There's just something about, like, putting, you know, large blocks of stuff into reality. Yeah. It's, like, very durable that jazzes me.

You know, I really, really like it. You know, now I build software, which I really enjoy. But, you know, I do miss the site. I do miss the construction. projects. I do miss, you know, the yelling and the concrete and all that stuff, right? Yeah. So it's cool. Yeah. Yeah. I mean, when we spoke two and a half years ago, we were kind of talking about that was like the ground floor of everything, right? You know, obviously not with our culture, but with technology. That was kind of the ground floor with things. So, you know, two and a half years, where are we at now?

Are we at floor one, floor two? I mean, you know, where do you think we're at in terms of especially with AI? I think construction is going through what manufacturing went through in the 70s. I think this is the most exciting period to be in construction probably the last two millennia. I think that when I started raising money for construction tech back in 2015, people were like, like, I had an investor that basically like listened to me for five minutes and said, hey, buddy, I don't want you to waste any time. I don't do construction tech. And so, forgive me, I'm like about to sneeze. But so, so what I was saying is that back in 2017-18, something changed.

It's almost like the VEC is huddled in the room and said, hey, we think this construction thing is going to be the next one that's going to like benefit. And I actually think there's many reasons for it. I think that simply put the machines, the reason construction is the second-leased industry is not because we're lazy. stupid is just that our industry is harder to digitize. Yeah. First thing they get digitized is finance. How hard is it to digitize numbers in an account, zipping from one account to the other? It's almost like the definition of digitization.

Right, right. Turns out when you want to digitize a building, and then you've got your architectural plans, your structural plans, your acoustics, your lighting, your estimate, your schedule, your specs, your codes, you're, you know, so many disparate, like, like, data sets. Right. So one, the size of the problem we're trying to crunch needed some really, you know, needed the machines to catch up. And I think they caught up really that 2015-16 period is when the machines started to be able to crunch data sets that size. Okay.

Moore's Law specifies, you know, the number of transistors in the microchip doubles every 18 months. Computers get twice as fast every 18 months. And they've been doing that. Most people think it's since the 1970s. It's actually 125 years that our species has been doubling its ability to compute every 18 months. months. Yeah. And if you look at that, we have been able to increase our computational ability as a species

by 10 million trillion times since the 1900s. That's not a lot of that. It's continuing. And so I think the other thing that you're seeing is with these LLMs that are coming out, right, you have the ability to access large data sets, right, large fragmented data sets, right, which you've not been able to do before. And that's basically what you're seeing with us today. And so it's just a very exciting time to be in construction right.

I think that you're seeing companies that are generating designs rapidly. You're seeing companies that are generating schedules. We are rapidly. You're seeing companies that are answering questions, you know, trunk tools, right? You literally ask a question or answer anything you want, right? You're seeing companies that, you know, high. part of doing design, right? Like, there's just lots of, you know, excuse me, like, cool shit that's, that's out there.

Right. Like at this point, it's not a question that isn't enough stuff. It's just how do you analyze it? Is this very similar to Silicon Valley back of the day where it was like the startups are, like you said, they're fast and furious. It's, they're coming online every day, it seems like. So will that parse itself out over the next five years or so? I mean, we will have too many at one point, right? Or do we need that many? No, I don't think so. I think the name of the game has always been the same, right?

The one thing to define Silicon Valley is failure. Most people don't realize that, but there's a lot of it. Oh, my God. Like, you know, I told you how many times I've, like, gone to see a friend who closed the company and they're freaking depressed. And, you know, I'm the one that's buying the drinks, right? And it's just, like, I don't think so. I think that you're going to see more of these startups.

You're going to see, like, exponentially more and more. more of them and they're going to fit. There's going to be one to succeed. And there's the next, you know, next version. Now it's the LLM craze, right? Just give it two years. There's going to be something else. Right.

And it's the next craze of whatever it is. I think that the industry needs, you know, we represents 10% of world GDP. Yeah. We need 10, 100 times many startups. Yeah. I really like, I see startups that are in the scheduling space. I'm friends with those guys.

Dev from Endplan, for example. I wish them the best. I hope they kill it. I hope they're a $10 billion business. I've never been competitive anybody. It's on a zero-sum game. There's enough for the pie to go around for everybody. There's a bunch of scheduling startups that are coming up.

Amazing. Like really, really happy for those guys and others. You know, I really like, I wish everybody, it's such an exciting period to be. I mean,

whether you're, the thing that's cool about this period is no matter who you are, you can play the game. If you're an exec at a company, you can play the game. If you're a scheduler, if you're a project engineer, if you're a journalist, if you're a podcast, you know, a host, if you're a researcher in a lab, if you're a student, if you're a CEO of a large construction. Like, whoever you are, you get to walk that wave. And there's, there's a role for you because we all need each other.

A startup doesn't exist with a lot of company that's paying it. It doesn't exist with route research that's coming out. Like, it's all an ecosystem. And it's, it's cool. just an exciting time to be in it. Yeah, there's a fine line of that success failure, right? So what do you think that secret sauce is? Is it implementation?

Is it, hey, you can create something great, but if no one knows how to, no one knows it exists, right? So there's all these factors that can play into it, or is there a lot of luck too? Jeremy, I think that like if you knew the answer to that, then I was looking for you to answer it, man. You know, I don't think, like I've seen

teams that were like the absolute top they would just blow you away right yeah just incredible caliber really just nailed the execution etc etc didn't succeed i've seen the opposite i've seen teams where i'm like i don't know right yeah you know the team's good maybe seers not good text not that great and they nailed it right they they weren't too early they weren't too late like timing is something you can't control right um yeah You know, there's always an element of fortune. Yeah. I think, like, I think to me, what I respect is people that have the guts to roll the dice in the first place.

Yeah. Because it's, it can be brutal. It can be nauseating, right? You try it. It doesn't work. Try something else. Like, by definition, innovation is, most people think innovation is like, oh, I have the spark and I, like, I've tried it. Like, it's never that way.

There's always, like, even if you have, like, the technology and it works. Now you've got to figure how to, you know, implement your processes. How do you run it? How do you do all that? all over these things about it, right? So like, yeah, there's just a lot of, it just takes a lot of, like, trial and error. Just persevereating, is the word I'm looking for.

Right. Yeah, I mean, we kind of spoke about it, but that, you know, last year we got a lot of media attention, especially with AI. It was all over the news, ChatsyPT, that just, all of this, it just, both personal and professionally. Do you think this is the year where we really take the shift from implementation, especially,

in a medium-sized large businesses. Is this the year where we kind of see more implementation of it? I don't think there's that big a change, truthfully. Like two, three years ago, I came out with, I had this moment and I came out on this statement, and I said the general contractor of the future will be successful based on three things. How good are they,

more efficient or effective, whatever, you know, where you want to use, at identifying, new technologies, evaluating new technologies, and integrating the technologies. And the reason that I stated out was that about five years ago, I kind of sat down. It was one of those philosophical evenings I was done with work. And I was sort of like, why do we exist as a startup? Like, it doesn't make sense. Like, sure, everybody in Silicon Valley tells you like, oh, we're smart and everybody.

We work hard and everybody else. I'm like, guys, I've been in Afghanistan. I know people that were smarter than me, hard of working than me, anywhere in the world. The thing is where I realized was startups are effectively the outsource R&D departments of large companies. That's why we exist. Right. What we do. And so the question to these large companies is how effective are you at leveraging these R&D departments that have been basically, you know, mushrooming, mushrooming up, right?

Right. And that's kind of like the thing. For us, you know, we had a major release six weeks ago. We released Alice Corp. Right. And like just to give you an idea of like what technology can do, we've invented the first AI scheduler in the world. You can go out there, schedule, like literally it takes a construction schedule

and then like generates millions of ways of building it. Parametric. Yeah. What VIN does to design, you can go to the height of the columns and it re-redraws your entire 3D more than else. Alex just for construction. You can add a crane, add a crew, try over time, we sequence, whatever it is.

The setup time used to be like in the order of weeks. 100 hours, 200 hours, right? We sat down, built a bunch of tech around it. Set up time is not one hour. Wow. Yeah. We can literally like, like my sales cycle has gone from I go to a customer and I say, back in the day, I was like, hey, I got this like great tech,

you're going to love it. Give me some time. Give me a few weeks. Give me two weeks. Give me four weeks. Give me six weeks. We're going to set it up. You're going to love it.

It's going to blow it. Not your socks off. Yeah. Come with me to the promised land. Yeah. And then like now I'm just like, I'm like in a meeting, you know, people are talking. I'm like, guys, guys, time on. Can we just, can we just, can we just, do you have a schedule?

Cool. Let me just, let me just import that and show you what it can do. And we literally import the schedule and boom, it starts running things, you know. Yeah, it's kind of mind-blowing. Yeah, I mean, and there's a lot of discussion about, you know, obviously we have a labor shortage. We really have for decades, you know, and then we have the baby movers that are going to be retiring. So we have their gap is going to get bigger, right, before it gets smaller. So how can AI help with that?

I mean, obviously we're going to have new jobs, right, in AI and how just like you, how do we implement this? But how can we be more efficient using it? AI so that we can replace some of those. You know, Jeremy, I'm going to tell you something that's kind of, I think it's almost sad. Okay. So everybody is always like, oh, it's going to take our jobs. We're all going to have no jobs, et cetera, et cetera, right?

Like, okay, great. And I've thought about, here's something that I spent some time thinking about. If you look at our species progression, let's call it since 1890, the Industrial Revolution. Okay. Name any metric. the cost of fresh water, the cost of sending an envelope across the Atlantic, the cost of traveling across the Atlantic, the cost of accounting, right, the cost of copying a document, the cost of buying a book, the cost of distributing a book, like name it, anything. Every single one of these things

has improved by an order of 100, 100, 10,000, whatever it is. Right. we as humans tend to view so so why are we still working 40 hours a week i don't want to what's that i don't want to but i do right but think about it if your job was an accountant and you were supposed to account for a certain company and somebody gave you um somebody gave you um a tool that enables you to do a job 10,000 times faster shouldn't you be working 30 minutes a week Absolutely. Yeah. Here's my answer. And this is the part that like, so I thought about it.

I was like, I actually think that unfortunately for us, our primary goal from an economic perspective is not producers. Our primary reason for existence is consumers. And that's the part that like kind of sad, right? I'm a guy that like invents technology. It's been all day thinking about patent this stuff. you know, I love it. Right. Like from an economic perspective, like our primary reason for existing is consumers.

And the reason that's significant is that technology is not going to change that equation. If you take away a million people's jobs, that's a million people less that need to buy stuff. And I think that's why that has been so resilient over the last 125 years. It has never result, it has never, not once resulted in us suddenly being like, oh, I got lost. a free time. Yeah, it's like we just replace it with something else, right? Like, okay. It's the same way it's going to happen.

Yeah. So that's that maybe, right? It'll be something. Well, now it's like, okay, well, you can manage 10 real estate projects, right, instead of one. Now you can manage, you know, a billion of construction, right? Like, it's just what you need to do will increase, right? Yeah, it does seem like the pace of our adoption,

and understanding. Obviously, it's just we are all understanding it more. I think in the beginning, I think AI was scary for a lot of folks because, you know, machine learning and there was a, you know, there's some people have personal feelings about it, but it does feel like now people are understanding. I mean, I know that most people that I know are using that to write their blogs or even the simplest ways of using AI to help them be more efficient. So it does feel like we're at least on that track of like, okay, how can I make this work for me to, I wish it would, replaced hours, it's not for me, like you said, it's just making me a little bit more efficient, I think.

I mean, one of the questions I often asked, like, here's what I teach at Stanford. But we teach Alice, we teach generative AI scheduling. The truth of that is, is like, what do we want to teach these, these, you know, young folk that's going to be useful for them 20, 30 years from now? Because someone's going to invent another Alice, there's going to be another version of it and so on and so forth. There's a couple of things that I can guarantee you are going to be useful. to you. The one thing I tell them is, hey, guys, you can buy an hour of Amazon EC2 compute for

two cents. If you are going to base your career on crunching numbers, I hope you're going to be satisfied with salary of two cents an hour. The thing that if you're going to be basing your career on managing the processors, the computers that are doing the computing, you will have a long and lucrative career. And the next question I asked them is the question I would ask myself was how many processors do I control? Because they're free, effectively. How good are you at basically leveraging that free sort of compute power that's out there? That's the question you've got to ask yourself.

Right. Right. You almost need somebody, if you can't do it yourself, you almost need somebody on staff that can, right? because you're just going to be missing out on all the benefits. Speaking of having people on staff, we've got a question coming from Deepak Mystery. And his question is, what's been your experience of engaging the leadership teams

within client organizations? Do you find it's still an uphill struggle in terms of educating them about the benefits or anything they switch on to embrace or experiment with new tools? I find internally with an organization, there's so much appetite on the shop floor, but convincing leaders can be quite challenging.

That's an interesting question. I don't know if I really differentiate. Like, I've been selling technology to construction for a decade. Mm-hmm. So I, like, when we started in 2014-15, there was, like, there was now such, like, people like tech. Like, we'd explain like, hey, run a pilot. They're like, what's a pilot?

Oh, pilot. Oh, I get it. Okay, that's how we test it. Right. Right. What I've found, actually, is I don't know if there's so much of a difference between shop and management. I just think it's, one, the type of organization. There's organizations out there that just, like, pride themselves, like, you know, DPR, Mortensen, Swinerton, right? Like, these are folks that are, like, always pushing the envelope, right? We work with companies in the Middle East that blew us away, right?

We're just, like, like, pushing the envelope, right? And they know, like, yes, we're going to try it, and it's not going to work. We're going to try two, three, four things, and then we're going to implement our processes, and it's a process. So I think the type of company and the type of person, And the people think, oh, the negative thing about innovation is that it takes time and there's failure to it. I think that's great. I think that means that it always is the great equalizer. It means that if you have the biggest market share, your number one in your industry, or your number one you field,

it always gives the opportunity to someone else to overtake or someone else to grow their career. Yeah. You can see that with the younger generation folks that are learning how to use Alice, folks that are learning how to use these new tools, immediately have a seat at the table and they bring someone at the table because they don't have the experience that other people's at the table have. Yeah, I mean, I think that what Deepak is saying too, though, is some of the older generation, they're not,

they don't see the technology as easily as the younger generations do, right? So they might be like, I don't know if I want to implement. That seems like a lot of work, right? Whereas the younger folks are like, no, no, no, no, let's try it. You know, this looks like fun. So it's like maybe there's a little bit of that. And obviously as they retire, it'll be replaced by younger folks. But maybe that is a challenge for a lot of businesses where the top down,

they're not seeing the benefit as easily as we are. Yeah, it could be. I mean, like I said, I think that it's always been the case, right? There's people that are willing to go through the pain. And there are what are called the innovators. And there's people that want to. not they want to be second. I've talked to companies that are like,

we're second movers. That's what we do. Yeah. We're not first, we don't want to do the bleeding edge stuff. We don't want to do cutting edge stuff. We want other people. And they're literally named.

Like, here are the four companies we look at once they figure it out. We're the ones to take it over and go from there. And that's fine. Yeah. There's nothing. There's nothing wrong with that. That's the strategy as well, right? Yeah.

Somebody, somebody out there is, is willing to take the risk and want to push further and try more. Right. Yeah, I mean, Alice has really gotten more heavily into the scheduling, right? So why is that such a tricky thing in construction? I know it seems very obvious that there's a lot of irons in the fire and a lot of balls in the air. And so why is that such a problem to solve?

And it's been a problem to solve since day one, right? Since we first started building something. Okay. Let me maybe tweak your question. Why has no one else solved this darn thing? I've wondered. I'm like, but the heck,

like, why are we the first ones in the world? I will tell you what I think the answer is. Scheduling is a high jump problem, not a long jump problem. So here's what I mean about that. There's certain types of problems. When we start our PhD,

literally the head of the department said, you guys need to pick a long jump problem. A long jump, when you do a long jump, you get that many points. Whatever you jump, you get that many points. With the high jump, you make it or you don't. And scheduling is one of those things. Think about it.

Let's say I have a scheduler. Our scheduler, it's a simulator that fundamentally simulates a construction project. And I go to you and I go, Jeremy, I got a simulator. And it works great. It's perfect. It does everything. And I literally had this conversation with DPR, I think, in 2013 or 14. I went to them, I was like, guys, we've been in the lab, like, we've fried all these brain circuits.

Like, we've really, really poured a heart into it. and we've got this, we've got this amazing tool. And the guy uses it and he goes, it doesn't do cranes. And I was like, okay, yeah, but it does everything else. So like, give it, it doesn't do cranes. Like, it's totally worthless. And I was like, but, but we've literally solved everything else. Yeah.

Literally. Precedents, equipment, label, production rates, calendars, like, literally the whole thing. And he goes, yeah, it doesn't do crazy. And it's useless. It's worthless. It's literally worth zero. You've solved 90, 95% of the problem technically, and it's worth zero without that last part.

It's all or nothing. And I think that's one fundamental, like, property of this problem why so many people have opted not to go for it, because it's an all or nothing game. And so, and it took us really, I mean, we had, you know, 10 people, three and a half years. Some of the, we have the top 25 mathematicians in India. Yeah.

You know, and he reached, he was like, I can't solve it. I was like, wait, you can't solve it, you know. But it's a difficult problem. It's a very, very difficult problem. We've cracked it. So what was the reason for the Alice Corps? What was the shift there or what was the change that was needed? We listened to our customer.

You know, yeah. Never do that. Never listen to your customer. I mean, it's basically like, we said we have this product. It's, it's amazing.

Like Jeremy, trust me, it's a, it's a Ferrari. It's amazing. You literally take a 3D model of a building. You set up constraints on it. You're like,

this is how you build a column, apply to all the columns. I build a slab, apply to all the slabs. And then you hit simulate and it literally shows you a video of the thing getting built. Like, you have two towers.

you have one crane it builds Tower A and then Tower B and you literally switch to two cranes and both towers go up the same time. It's like, it's like, I mean, to me it was like watching something think. And I remember the first time it outsmarted me.

But to answer a question, why core? What we didn't know then was was it possible to set up constraints visually for a construction project on a 3D? That was like, Like, you know, people like absolutely not. There's no way you can frame to do that.

Right. There's two downsides. Set up time. Took a long time to set up, but it wasn't integrated with an existing schedule. So it's really useful when you don't have a schedule to start with it. You're trying to figure out, like, what is the overall sequence? But let's say you're dealing with a construction project that's in flight. They already have a schedule.

They don't want a new one. Right. So we sat down and we said, okay, well, we did the impossible wants, which was we figured out how to convert construction to algorithmic terms. So could we do the impossible squared? Could we figure out how to take all that optimization juice and apply it directly to existing systems? Yeah.

Would it be possible to take an existing schedule, which is task-based, and then apply the constraint-based, element-based, AI-based universe that we've built to it. Right. And that's, you know, 18 months ago, we sat down. In January of last year, we had started drafting up prototypes and working on it. We did not know if it was possible. Really.

We're like, I don't know if we're going to pull it off. I don't know if it's doable. I don't know if it's not doable. We released it. We started working with our beta customers in November, released in February. the thing that blew me away by it is not only is it possible, but we can basically do everything we were able to do with the 3D system. There's nothing that you've lost in this approach,

which is kind of crazy. Like you could take an existing schedule, import it into Alice, and press the button, boom. It works best with resource-loaded schedules, but you can basically allocate resourcing to it in like 20 minutes. You can do any, like, the max, X in the extent, if you wanted to run everything we can think of is like three and a half, four hours. It's crazy.

We were, literally, we had a large project we were looking at in the Middle East. And we're like, hey, guys, send us a schedule within an hour of around 35 different simulations for them. They were worried, for example, they were worried about the swings in the cruise. You got 1,200 people on the day. You got 600 weeks later, 800, 600, 1,600, 1,000, 1,000, 1,000. literally like we're like type that type simulate like literally the graph goes up to 600 stays at 600 project ends like yeah wow that is crazy it looks like has got some questions there on lincoln you can scan that real quick let's see what is he looking for tracking r oa data for software implementation

yeah this is difficult there's there's a few cool thoughts Senator Hassan, so thanks for asking. It does bowl down to culture, right? I can tell you that a certain company is either we're on the bleeding edge, we're on the cutting edge, or first movers, second movers. We're laggards. We wait for everybody to figure out. Our competitive advantage is our legal friend.

Our competitive advantage is that, you know, we have people in Washington, D.C. that give us government projects. Our competitive advantage is we have a 50-year relationship with DOT. like there are other ways other than technology right um oftentimes the technological advantage tends to catch up and like sort of beat a lot of the other ones right but yeah it is about culture um in terms of looking for the next thing i totally agree with you hasson right i i i get stressed when i go to talk to a chief innovation officer or someone at a company and they're going like oh my god that's so exciting there's like so many cool things out there like we're going to try all

of it. Versus, you know, we sat down. We looked at our, we have a process map. That's probably a good place to start. Here's how we manage projects. We go through bidding. We go to pre-construction. We go through construction.

In pre-construction, here are the steps that we do. We look at the estimate. We look at the schedule. And when we sort of drew out how we do business, we thought that the highest leverage, the technology readiness, versus the advantage we get, right, are going to be these three areas. We think that we need to apply to scheduling.

We think it to apply to these are the three areas that we're going to apply to. That's what we're going to do. Right. Right. And so to answer the question is, like, once you have a strategy that is based not on just external stuff that's happening, but internally in what you're doing, that's probably a way to go. And the last point that Asson has it on terms of ROI, we have found that ROI calculation, like, a failure to us is when we go to someone and we run it and everyone's like, oh my God, that was amazing.

Then I know that we failed. Because then they're like, well, okay, well, what was amazing? That was great. It's the future. There's our people are going to schedule. Okay, well, how much money is it? did it make you? Okay, well, what did we save?

And so for us, we're really, really diligent and stringent when we go to someone to sit down and say, guys, we want a data-driven approach. The technology really works, but we need to sit down. You guys got to let us know how you're measuring this. The number of schedules we created, is it the savings that we saved? Is it the delays that we've mitigated? Like, what are all the different ways that we can save you guys time? Right.

Right. Or money. Or what is it that you're from, mitigate risk or connect with your estimate or whatever those is. And so, Hassan, yeah, I think the two points, like, culturally, you've got to be the company is pushing that innovation or at least, you know, supports it. And then you need a strategy of here's what I want to innovate on. And then you can be able to measure how those strategies, you know, those innovations are doing. Because part of innovation is also killing stuff that's not working.

Right. Right. Yeah. I mean, like you said, though, having a strategy in place as a team, like,

hey, these things will be best for us, as opposed to what Hassan said is like going after this shiny thing every time. And usually that's not going to work out too well for you, right? Because sometimes the shiny thing isn't ready for you yet, or it's not ready for you. So,

you know, he's right. You got to really put some thought into it. before you just go after the next best thing. Yeah. So tell me a little bit about this. Alice joined the Oracle Partner Network. I know that historically you guys were maybe considered competitors.

So tell me a little bit about that. I was kind of curious about that partnership. I mean, Jeremy, like I said, I really don't know if I believe in the zero-sum competitive game. Yeah. If you look at some like Oracle, they are one of the biggest companies around, like focal. They have, you know,

biggest market share for construction scheduling also globally, right? I don't think that a company like ours with, you know, 80 employees is competing against Oracle. Right. They've been, they were the first, you know, the Primavera, their product, you know, they bought it, but Primavera was the first scheduling, so like, let's be fair.

When Primerra came out with this technology, in 1980, I think it was 83 or 84, I forget right now, but it was early 80s. That was a huge leap forward. That was a huge step forward for the industry. And let's hand it to them like they are the reason that there's some scheduling technology that's available for the construction industry. That's amazing. At Alice, we've rightfully, as the new generation, right, built the next version of what, you know,

P6 was doing, right? We've built AI scheduling algorithms. And by the way, 20 years from now, there's going to be another PhD sitting somewhere in some like university. It's going to be like, ah, like here's the next, you know, thing that the schedule needs to do. Right. Right.

That's how we're moving forward. Right. No doubt. What segments of the industry are, do you guys, does Alice focus in on? We focus, what's really interesting about core is we used to do projects that are 100 million and up. Now I do 10 million up. Right.

It really does. Like the setup time is like one hour, two hours. It's crazy. We tend to look at infrastructure industrial. Those are two of our sweet spots, right? We tend to, the reason is that ultimately what Alice does, if you want to optimize something, you need to be able to like shuffle resources or squeeze out resource efficiency or squeeze out utilization, right? Reduce swings, like change production.

All of that stuff requires you to have some production model. And if your answer is, well, I don't know, I just subcontract everything. I don't really worry about how to build it. I'll let the subs do that. Probably, you know, Alice is not the tool for you. We have ways that what we do is we optimize durations instead of resources. We definitely have started looking at doing that. And there's actually a really easy way to do what we call parametrized durations.

literally you take the project, you attach something that we call labor hours. It's a resource called labor hours to your projects. And then you hit schedule. And really interestingly, you can start doing some really interesting analyses on how many people do you have on site. What are the, can you tweak the durations by moving resources between tasks? And like, can you increase or decrease durations by call it 10, 20%, or even double or triple, you know? and then that starts giving you insight into what is possible in terms of affecting the critical path, moving activities onto or off the critical path, having more than one critical path, changing the critical path, increasing, decreasing flow. Like, that's the kind of stuff you can do very rapidly.

But again, like, you need to be someone that actually cares about analyzing the schedule. Sure. And you laugh, but not like, we have talked to companies that are just like, no. Literally, they're like absolutely not. We basically hire subs. We basically tell them, this is when you start, this is when you end, and that's all we do. And we're like, guys, you're building like a multi-billion dollar facility. It might be worth literally four hours of a time to let go and optimize this stuff.

But everybody's got their own way doing it. Interesting. And your team was saying that in the UK, especially in Europe and especially the UK, that the adoption's been a little bit quicker. Why do you think that is? That's kind of wild to me. Yeah, I think there's two reasons. One, I think the Europeans are extremely process-oriented.

Like, look at the EU. Okay. It's funny how, you know, oftentimes everybody's greatest strength is their greatest weakness. True. You know, and in the EU, I mean, you know, it's a hierarchy. It's a process. You know, look at the European Union, right?

like anybody in the EU drives them nuts. Everything has a protocol and a standard and everything is top down and so and so forth. At the same time, like, you know, the UK government said everything, everyone was going to use BIM, and that was the end of the story and everybody uses BIT. And so I think that because they're process oriented, because they actually like are diligent in the way that they manage their projects, when you give them a software that leverages that, that actually makes them happy. Yeah.

Yeah, strange how that works sometimes. Yeah. And also, you mentioned before that you sell to both GCs and owners. How do both of those segments, how do they use it differently? Really, really great question. The thing that's really changed about core, so we, our product used to be, now we call Alice Pro, the 3D Pro. Alice Pro was you import a 3D model. You attach constraints to it and you hit simile.

Okay. If you were an owner, think about this. So as an owner, I'm going to go. I'm going to say, okay, great. I'm going to import the BIM or I'm going to create the BIM or the ALICE team will create a BIM for me. Usually the BIM is like a crude version. It's not a detailed bit.

Right. Like, we'll do 200. Just give me the shape of what you're building. Sure. Do that, well, less than a day, even a billion-dollar project, maybe two days. Create that, put in the system running. The challenge there was that the schedule you had wasn't a one-to-one match. with a GC schedule.

Our schedule was usually more detailed, 6,000, 8,000 tasks, GC schedule was 2,000, 3,000 tasks. So then you're like, okay, well, how do you compare the two? You know, we have like translators and so on can still sort of. With Core, as an owner, you're literally like, hey, can you send me a schedule? Awesome.

Within five minutes, I can do what we call schedule assessment. Let me check if the number of resources that you're telling me are actually the resources I need. Is there any resource over allocation? What happens if I schedule it with the resources that I have given? What happens if I schedule with the resources in the P6? Those questions, by the way, within like 10 minutes you can answer. I was working with the lead scheduler at one of the large infrastructure companies in the U.S.

We've been working with them. They're kind of all over our website, lead schedule. He said, look, Renee, the P6 schedule says that this project is now, let's say, six months late. And I don't think it's true. I think it's worse. But I can't prove it. Within 20 minutes, Alice pops out the answer that he was suspected, which is, yeah, this is how late we are. Now what you do?

How do you bring that back? From an owner perspective, the fact that you're living in the universe of the G.C. is a huge deal. It's a game changer. Right. In fact that you can actually do this stuff within an hour or two is also a huge deal. Yeah, no doubt, no doubt. What other aspects of construction is AI being kind of implemented right now?

I know scheduling is the biggest one, but what other things you find that are promising? Construction is large disparate data sets. And so what the latest craze of construction, which is the LLMs are doing, is they're able to answer large disparate datasets. and companies like trunk tools or PARSEC, right, trunks tool's program IO, enable you to actually look at large data sets and in the first case,

just answer any question you want in your project. And in the second case, literally match your lighting, lights to the actual light that will satisfy with the manufacturer. This is the order for this project in order to like satisfy that. Right. Those are two like applications directly off the bat that I think the large fragmented

data set analysis is something that I think is kind of a cutting edge stuff that's happening today. Yeah, right on. I think Deepak has another good question here. He says, what's your take about traditional roles within construction industry are being, or will be disrupted by AI? Are boundaries between roles being blurred, planner, risk manager, cost estimator, project manager?

Do you think new roles will emerge? If so, what kinds? Great question. It's a great question, Deepak. It's already happening. Yeah. So when I started in 2015, there was no such thing as a chief innovation officer.

Like, we would hear of like, head of VDC, virtual design construction, or head of bin. That was it, right? Yeah. I think that now chief innovation officer. What I'm also seeing is a number of companies, companies were talking to have data scientists.

Whoa. That was not the case even four or five years ago. And when I say that, I mean like that's kind of let's say, I don't know, if you look at the ENR 400, probably, I don't know, top 20 of them have something like that, but, you know, what are we going to do with the data? So data scientist is a new one. Prompt engineer is a new one. So to answer a question, Deepak, it is happening. Right.

It's not like is it going to change. It is definitely happening. If you're looking at companies, Dustin Devon just found it something. He's the guy that founded Building Connected. He just found it something that helps estimate us, right? So the type of work that the estimator will do will change, the type of work the schedule will do will change. Like with Alice, right?

Like you're suddenly as an estimator back in the day, it was like produce a schedule, you know, it'll help me God that will basically satisfy the owner. That's all I really want you to do. it's like, oh, can you produce the optimal schedule? Can you produce schedules that are mitigating delays? Can you produce resilient schedules? So that's the kind of stuff that you're seeing.

Are we going to see the elimination of like, oh, you don't need a scheduler and estimate anymore? I'll tell you a secret, Deepak. When we found out of the company in 2013, 14, like I was still doing my PhD, but I was like, oh, we're going to get to a point where you're like totally eliminate the scheduler. right like who needs a scheduler boy that i like chew those words like abs of freaking ludley not no you are not eliminating that person like this computer is really good at crunching lots of data awesome right

socks at gut sense um like it you still need some like you know what's funny people don't select the fastest cheapest schedule almost never because there's too much risk because the I know something about that project in terms of the number of electricians that are available. I know something like the gut sense of risk analysis for understanding

reality, right? It's just not something machines can do even with LLMs. They can write something for you, but they don't understand the value of it or the context how it fits in the global picture. And so you are not eliminating in the human

in any in the near term. You are definitely limiting a lot of the slug work. And maybe in the short term, yeah, you'll be like, oh, well, I don't need three estimates. There's any two, but I will guarantee you, if you look at the last 125 years, the only thing that's going to happen is you're going to go back to hiring those three estimators instead of estimating 20 jobs. Now they're doing 200. Yeah. It's literally been this, it's been 125 years of this stuff. Right. Yeah. Yeah. No, that's a good take on that. That's a good analysis. So,

Crystal Ball Time, where do you see AI going in next five years, 10 years? I know that this is like maybe even be hard for your brilliant mind to figure out. Is it beyond us? Like, what are you seeing this thing going? If you look at that document of Kurzweil's law of accelerating returns, he predicts that you would buy a human brain equivalent of compute for $1,000 in 2023. Chad GBT, you know, comes out, breaks through. in 2023, kind of crazy.

Yeah. He predicts that you'll be able to buy one human race equivalent of human compute in 2049. Wow. So that's a data point. Another data point I'd give you is, personally what I think you're going to start seeing is composite AI. Right.

So I've got a friend of my called Philip Vosak, Vodak and American. He runs a company called Philuta, and they're AI. That's what they do. But composite AI is basically when you start combining. You can start combining LLMs with discrete event simulation with stuff like the combination of layers of AI, which is, by the way, how it's worked for 20, for 25 years.

Any, any, Alice, you know, uses, you know, I have no, six, seven, eight different types of AI to do what it does. And so what I think you're going to see is a combination of various AI. Like one thing with the LLMs don't do very well is math and engineering. I'm fascinated, fascinated to see the combination of the old school engineering and math kind of AI deterministic equation number crunching with this new world sort of neural network stuff and how that's going to play out. I think there's going to be a number of companies that come out with as I call it more to use Philip, you know, Phillips terms as composite AI approach.

That's what I think you're going to start doing. Okay. Yeah, that makes sense. what about personally do you is there anything that that scares you about AI is there you know is a regulation that you think we need personally or how do you feel about the personal aspects you know you know mirror link robots all that kind of stuff what what do you feel i don't think like the EU just came out as as the EU does i'm half check so i get to like ding the EU a little bit but like you just came out with regulations for AI right and i'm thinking to myself guys like

Bravo, right? How about maybe inventing some AI, right? I don't know how effective the regulation is going to be because I've got some news for everybody, which is the cost of producing open AI's LLN is like in the vicinity of $30, 40 million. It's, that's not an insane amount. Like you don't need governments to do that. You don't like, like, There are many companies, even individuals that could do that. Sure. So what I'm trying to explain is that your ability, the cost to produce these things. Like, there's kids in a lab somewhere that are building stuff that would probably scare you. So, like, I don't know if regulation is really the answer, right? I think it's just letting the technology do its thing. Am I scared?

No, I'm not. You know, like I'm a technologist. I really believe in the ability of technology to continue doing good for humanity. Right. I think that, you know, will we reach a point where we will build something that is more effective than our species? Like heads up, guys, if we want to travel between stars, you know, you need something that can, you know, be alive for a million years or whatever it is to get to the next star. Right. You know, I think that there are things that humans are good at that we will continue to be good at for the next at least 10, 20 years.

so I'm not too concerned about like AI taking over the world. I don't know. I don't worry about it too much. Yeah. And I don't think regulation is the way to go. I simply put because I think it's moving so fast that like who's going to regulate it, the government?

Like, you ever watch Congress have a conversation about AI? Like literally have you watched the interviews they had with Mark Zuckerberg? I love it. Yeah, that was crazy. There was a couple of questions where like Mark was sort of like, uh, like are you seriously at? like like you know so i i don't think government regulation is the way to go i don't think it's

much of a difference right right yeah no well thank you so much for your insight rene um how do people get in touch with you um is it linked in or do you go to alice so tell us a little bit about how to connect with you uh alice technologies dot com okay form i will happen to get in touch with you shoot me an email rene at alice technology dot com i love talking to construction geeks um if you've got a project you won't schedule, if you've got something that you think can't be done, you know, shoot us, shoot us an XCR file and we'll take it for a troll for you. Yeah, awesome. Well, thank you so much for being here.

Your insights are amazing. It's good to connect with you to kind of see where we're at and where we're headed. So thanks so much for being here. Jeremy, always a pleasure. Yeah, we'll see you in two and a half years. Yeah, that's right. Thanks again for being here on Construction Executives Live. I'm your host, Jeremy Owens.

Hey, thanks so much for tuning in. I see a lot of familiar names and faces. Thanks also for the chatterers for tuning in and for giving their insight. It's always helpful to kind of to have people chime in a little bit. We'll see you guys next month for another show. In the meantime, check out build12.com. Check out alice Technologies.com. Check out all these great folks that are doing awesome things in our industry.

Let's support those who are pushing our industry forward. So thanks again for being here. you guys next time. You've been listening to In The Zone and Construction Executives Live with Jeremy Owens. Be sure to subscribe to In The Zone and stay in the know with the best minds in the construction industry. To nominate an innovator or change maker in the construction industry, connect with your management peers and stay up to date with construction industry news.

Be sure to visit usconstructionzone.com.