The Risky Planner™

Prediction Scorecard: Capital Project Forecasts vs. 2026 Reality | The Risky Planner S2E18

Albert & Nate w/Dokainish & Company Season 2 Episode 18

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 1:01:37

Send a text

A year ago on The Risky Planner, Albert Brier and Nate Habermeyer made a series of predictions about where capital projects were headed.

Mining electrification. AI adoption in project controls. Autonomous equipment risk. Data center energy. Nuclear deals. Mega project cost performance.

This episode puts each prediction on trial against sourced data from 2025 and early 2026.

What you will learn:

00:00 Cold Open

03:00 News: Belt and Road Initiative hits $213.5B in 2025 construction deals

06:00 China's energy advantage and SMR race

12:00 Mine electrification: market tripling to $10.51B by 2033

15:45 Cogeneration: mining companies become power generators

17:00 Grid stability as a scheduling dependency

22:00 Mining vs. AI: competing for energy

26:00 Autonomous haul trucks: near-zero incidents with fleet separation

33:00 AI adoption: 12% usage, 29% unprepared, 40% price increase by 2027

39:00 AI job displacement: 37% of US companies replacing roles

43:00 Nuclear for AI: Meta signs 6GW deal in January 2026

44:00 Data center growth: 14% CAGR, $3T investment by 2030

48:00 Fiber optic demand: AI data centers need 36x more fiber

51:00 Mega projects: 9 out of 10 exceed budget

54:00 California High Speed Rail: 59% of segment complete

57:00 Advice for project executives on AI-driven planning


Key stats from the episode:

- Mining equipment electrification: $3.05B to $10.51B by 2033

- Each battery electric vehicle: 600 tonnes CO2 reduction/year, 20% productivity increase

- Texas power requests nearly quadrupled in 2025, 73% from AI data centers

- Meta: multi-gigawatt nuclear deals signed January 2026

- AI data centers: 36x more fiber than traditional builds

- 109 of 302 AI models had price changes in January 2026

- Rail projects: 44.7% average cost overrun

- 9 out of 10 mega projects exceed budget


The Risky Planner is produced by Dokainish & Company.

Music by Thompson Egbo-Egbo: egbomusic.com


Presented by Dokainish & Company www.dokainish.com

The Risky Planner podcast delivers expert insights on project controls, capital project management, and strategic planning for today's complex business environment. Subscribe for regular episodes featuring industry leaders and practical advice.

Albert Brier:

Well, hi Albert, hey buddy. How you doing? Nate, been a minute? Yeah. Well, we've been, I don't know, busy. I've been working, doing actual work.

Nate Habermeyer, APR:

Work, getting paid to talk, getting paid to plan, getting paid to de risk. Oh, yeah, we brought on a new client, which was exciting. You've been you've been busy?

Albert Brier:

Yeah, I've been busy. What's really funny is we, I live in Alberta, specifically Calgary, and we brought on a new client whose headquarters are in Calgary, so naturally, the first thing they asked of me was to fly to Toronto to attend a workshop right across the street from our headquarters. I found that to be pretty amusing.

Nate Habermeyer, APR:

Well, I mean, we, we had the pleasure of you being in the office, so everything worked out in the end, right?

Albert Brier:

I like to Swan around and pretend like I'm important. Every time I'm in Toronto,

Nate Habermeyer, APR:

Swan around. I've never Swan around. That must be yours. Is that like a Texas thing?

Albert Brier:

No, it's definitely not a Texas thing. I picked that up from the Brits that I have worked with over the years. So I presume it's a British thing. But I guess I don't really know.

Nate Habermeyer, APR:

Sounds like a conspiracy, but you know what? Speaking of conspiracies, I've got something to check in with you about, oh, God, is this and you're gonna get annoyed. Okay, it's the Nate. Nate Dave, okay, you remember this is, like, going back more than a year. You're like, I don't believe it. It's not true. No, it's true, and I continue to prove it.

Unknown:

How's your how's your group going? How's your group going?

Nate Habermeyer, APR:

I unfortunately, it's just me. I'm the only member on LinkedIn, and

Unknown:

no weird Anyway, do please continue with your update, though.

Nate Habermeyer, APR:

Yeah, Nate. So remember, all you Nate's out there that are getting called Dave, go to Nate dot Dave on LinkedIn and join my group, one of the influences. So this guy, Nate B Jones, he's an AI influencer. He's on sub stack. He's one of my subscribers. I subscribe to his channel. He's got 102 followers. I don't know how many subscribers, but it's really great content. I mean, he goes really deep on all things AI. He's also a Seahawks fan, so, you know, congratulations. Oh yeah, all right, go Hawks. I was, you know, grew up in across the Puget from Seattle, so I was a suffering Seahawks fan myself for many, many years when I was younger. But yeah, anyway, Pete

Albert Brier:

Carroll showed up and made everything better, everything better.

Nate Habermeyer, APR:

So I reached out to him, and I just asked him. I was like, hey, you know, when you introduce yourself, do people mishear you and then go Dave? And he actually responded to me, and he was like, yeah, all the time, there's another Nate. And then randomly, I get some, you know, account executive sales guy that reaches out to me on LinkedIn. His name is Nate. His name is Nate, and actually, and so I said, Hey, I'm not interested in what you're selling, but I got a question for you that's random. Okay, fine. So then I asked him, I'm like, do you get do you go by Nate? And he's like, yes. And I'm like, Okay, I have this theory, like, do you ever get Dave when you were introducing and he said, yes as well. All the time,

Albert Brier:

all these conversations that you're having with other Nates, they're all happening without me or anyone else to validate that they're actually happening.

Nate Habermeyer, APR:

A new angle to the conspiracy is, yeah,

Albert Brier:

exactly like this is one of those two sided coins conspiracy situations. You know, the one side there's the actual Nate Dave thing, and on the other side there's you pathologically lying about it. At least that's what I presume.

Nate Habermeyer, APR:

But you have to know, you have to realize that Nate are, we are an innocent kind. We are. We are truly, no, I'm just kidding.

Albert Brier:

I'm just sitting here racking my brain for a bad mate, yes,

Nate Habermeyer, APR:

like there are none. Not the Epstein files nowhere to be

Unknown:

found. God, you really want to go talk about Jeffrey Epstein.

Nate Habermeyer, APR:

Actually, I do project. Can we pivot real quick? Just so I want to mention the listener survey. We do have some responses, which is really exciting. We're going to start promoting a little bit more, just as an FYI to run everybody, you know, visit our when you when you're going to our podcast, and you see, you know, go to the show notes, there's a link to our listener survey. Really like to get to know you better. So, yeah, we're gonna try and take your questions and we'll work some of your feedback into future episodes and whatnot, but the survey is going to help us do that for the rest of the year. So I'm pretty excited, like, for sure, actual listeners, but yeah, my neighbor, Dave, has submitted a survey. He told me, well,

Albert Brier:

a bunch of people who work for us submitted like test surveys to make sure that it was working. So I think, I think that those are still in the in the data set as well. So we had to do a little bit of culling before we really do Yeah, but which means that we haven't really reached a critical. Nate, we have a bunch of responses, but, you know, we need to go through them and make sure that they're all valid. So the more we get, the easier that will be. So, you know, do please get in there and answer the questions. We're kind of a big deal, right? Well, yeah, totally, a huge deal, yeah. I mean, we're low key, the biggest celebrities that you've never heard of, yeah.

Nate Habermeyer, APR:

No kidding. Well, so the news of the day, like, in terms of building, which was pretty interesting, I so Lunar New Year's right? Chinese New Year, it's like the thing. So a lot of you know Chinese people of Chinese ancestry or immigrants live in my neighborhood, and, you know, so we Gung hay fat Choi or Lunar New Year's, if you're from, you know, somewhere other than China and Asia. And it got me thinking like the Belt and Road Initiative. Like, what is that doing? In fact, like 2025, was a banner year for the Belt and Road Initiative. They spent, I would say a modest $213 billion on 350 deals.

Unknown:

Okay, yeah, you know, I guess they could have gone, they could have gone real hard when push it up to 300 if they're really trying.

Nate Habermeyer, APR:

But the interesting thing about China is, like we have talked about tariffs, we talked about Trump, we talked about their effect on projects in the past. And so I think what Trump is doing with tariffs and whatnot, it's actually like it's not isolating China, which is what the US wanted to do. It's actually creating, you know, it's helping China reintegrate. Like even Canada is turning towards China.

Albert Brier:

I Well, China's trying to buy for, like,

Nate Habermeyer, APR:

our new Chinese overlords, yeah, yeah.

Albert Brier:

You know, that doesn't seem to be what's happening. And it's particularly true if you're in the eastern hemisphere. But like, China is building what in the mid 20th century would have been called, like a co prosperity sphere. You know, exactly these, these giant trade networks. And, you know, calling them trade networks implies a bilateral nature to the trade that doesn't really exist in this case, like it's very one sided. You know, China is the economic powerhouse of the region, and they know it, and they've parlayed a lot of the American manufacturing that has happened there over the past few decades into some truly just second to none tech capabilities, like the way that they can churn out high tech solutions to everyday problems is, you know, it's the same way that people used to talk about Japan in the 80s. You know, where, like, they've got this future tech stuff, they're all just walking around with all the time. And, you know, they're, they're building all these things because we're paying them to and, like, that's exactly what's happening in China. And the level of of, like, technological growth there is just absurd. I mean, you know, the the dark factories and the proliferation of electric vehicles, and, you know, they're, they're building, they're building more

Nate Habermeyer, APR:

even, dancing robots. Did you see the dancing robots? I've seen dancing robots, human

Albert Brier:

or, sorry, dancing robots was cool. Okay,

Nate Habermeyer, APR:

right? Like, I think Wall Street Journal talked about that, like, like yesterday, and it was basically like, there is not a market for dancing humanoids. And I'm like,

Albert Brier:

Why do you feel going to say this, according to him? But also, like, you know, Beck had dancing as ammo robots in in the video for hell yes, like 20 years ago. So, you know, let's not forget about the hell yes,

Nate Habermeyer, APR:

years ago. Do you you mentioned Japan? Do you remember when you got your first Walkman?

Unknown:

Yeah, I do yeah. I was, I do

Albert Brier:

that, listening to listening to Pantera while mowing the lawn, absolutely

Nate Habermeyer, APR:

I was listening to faith, no more, okay, epic.

Unknown:

Were you mowing the lawn? Though? No, I was

Nate Habermeyer, APR:

probably riding a bike without a helmet. Though,

Unknown:

yeah, classic, classic, classic, Walkman behavior, you know,

Nate Habermeyer, APR:

ride a bike, no helmet, Walkman blasting and no reflectors.

Albert Brier:

Oh, yeah, no, 100% I mean, you know, my my unsafe Walkman behavior, yeah, mowing the lawn probably pretty safe, all things considered, but I definitely did light a bunch of GI Joes on fire while I was listening to stuff on my Walkman. So, I mean, that was probably not the best choice, but, yeah, but

Nate Habermeyer, APR:

it was cool. And speaking of, speaking of China, though, the the energy mismatch, well, the advantage, I guess, actually here, let me back up. There's a few things that I'm really excited about. Like, there are a bunch of llms, a bunch of models like Kimmy queen that are coming out of China that are really exciting, right? And you know, agents like open claw, that's from the US, but you can use the Chinese models that have huge context windows. So that's pretty exciting. But then their power advantage, like they have more energy production. More than 50% of the US right now. And I think the US is playing catch up, like, you know, all the SMR stuff, like, Canada is building right? Us is building like crazy. Like, ordering, they're building like crazy. Yeah.

Albert Brier:

Well, that's true. Look and by the way, just quick preview, we haven't gotten there yet, but we're going to be touching base with some of the predictions and kind of quantifiable statements we made last season later on in this episode. But this, this is one where I feel like we haven't really seen enough information to really know what's actually going on yet, because, like, you can place an order all you want to, like, either you send an email to a guy and say, I want one of your SMRs. I thought it's a little bit more control than that, but still bit of a black box. It is like, nothing's come out of that box yet. So when they start generating power, you know, call me and

Nate Habermeyer, APR:

it's like, it's like, when Mark Zuckerberg said, like, didn't there was that, that meeting where Trump asked him how much he was gonna invest, he's like, Oh, I don't know, 500 billion? Yeah, sure. Remember this like

Albert Brier:

just throwing numbers around, you know, just what's going to be a number that's big enough to appease dear leader, but not so big that I can't actually do it if the rubber meets the road, but it never will, so it doesn't matter. But anyway, back to China. You're talking about SMRs, but I think the real news there is just a nuclear growth in general. They have more nuclear capacity under construction right now than the entire rest of the world combined. And it's not even close,

Nate Habermeyer, APR:

not even close. It's insane. We're gonna start kind of going back to the one of the first episodes that we recorded, which was at the what do we call it? The Disneyland for adults, yeah. Disneyland, yeah. Mine Expo. It was so cool. It was it was so cool. And the scale of technology, the like advancements, the autonomous, the electrification, which electrification is what we're gonna talk about. But let me read the prediction to you, Albert, and then I think we can sort of dive into it. How does that sound? Sounds good to me. Electrification is clearly an opportunity for mine sites to conserve on diesel fuel and clean up their mine sites.

Albert Brier:

Well, let's talk about mine sites, cleaning up their mine sites. Yeah. So it looks like this is happening, right? Like we're getting rapid acceleration in the electrification of mine sites. So the market for electrification equipment is expected to triple in the next few years. So that's pretty cool. Battery electric. So I have a one of our new clients actually is in the strip mining business, and they have a bunch of autonomous haul trucks, which we'll get to later, but none of them are electric, right? So I'm curious to know if that's on. There they are. Yeah, they are okay. But a lot of the new automated ones are either hybrid or full electric. And there's, you know, a lot of investment being announced there, like this, mines in Quebec recently announced 40 million worth of investment in battery electric haul trucks. So each battery electric vehicle apparently reduces CO two emissions by 600 tons per year, and it also increases productivity. Presumably, that's because a lot of the electrics that we saw at mine Expo had a continuous feed, as opposed to a battery. So, I mean, they have, in addition to a battery, right? Like, they've got a, like, a rail or something that they're that they're traveling along,

Nate Habermeyer, APR:

and it's instant torque, right? Like, you don't lose anything with, yeah,

Albert Brier:

like, if you've ever met a haul truck, they don't really have a torque problem, but yes, like,

Nate Habermeyer, APR:

they could always use a little bit more, is what I'm saying.

Albert Brier:

Oh, sure. Like, who couldn't, right? Like, when we start drag racing these things, that's gonna get really exciting. But the main thing I think about this is the efficiency increase. Because if you do have them on rail, right, and they can, you know, maybe slowly travel between rails and in storage and things like that, like, they don't have to spend as much. Have to spend as much time. You know, in downtime, it's pretty much just maintenance. And maintenance for electric vehicles is way, way simpler than it is for internal combustion engine driven vehicles. So like, as the fleets get electrified, they're also going to become more operationally efficient, which is cool. So wouldn't you that's all

Nate Habermeyer, APR:

happy if you're regarding our strip mining client. So if you're kind of working on the schedule, like, put it into sort of the project management, you know, risk management. Like, what kind of risks would would we be avoiding? You know, Diesel versus electric? Like, how would you think about

Albert Brier:

that, the risk of a haul truck just giving up, you know, and not working well is a lot lower in this case, I would assume, Oh yeah. I mean, because the number, you know, oftentimes you'll hear about machinery in terms of how many moving parts it has, one moving part, you know, or a million moving parts, like electrics, are way more on the one side. Of the spectrum than they are, you know, whereas, like, the diesel stuff is way on the other side of the spectrum, right? Like, they've got zillions of moving parts, or really, really complex machines, so any single part of the like, they're really mature, right? Like, there's a lot of maturity around how they are maintained, and, you know, they're pretty reliable, and that's all wonderful, but you still can't overcome the fact that there's just a lot that can go wrong in an internal combustion driven vehicle, and a lot less in electric ones. So I think that those those risks go down for like, project work. It doesn't make a huge difference because

Nate Habermeyer, APR:

it's the maintenance. It's, yeah, you can

Albert Brier:

maintain a smaller fleet that's more productive and all that stuff. But let's, let's pivot. Another thing we talked about at mine Expo was cogeneration.

Nate Habermeyer, APR:

This is interesting. There was some new, like, there's, there's new stuff that's coming online, up on the, you know, the iron is it? The Iron Shield, you know, like around Sudbury, lot of mines in northern Ontario, Faris

Albert Brier:

is gonna kill us. I don't remember what that's called, but like, yeah, that area,

Nate Habermeyer, APR:

yeah, up north, up north.

Unknown:

It's up, it's up. Where people, I guess, live. I'm gonna cut Ferris is,

Nate Habermeyer, APR:

like, close to the close to the Arctic Circle, I think, right,

Albert Brier:

isn't that? No, he lives in igloo

Nate Habermeyer, APR:

in Toronto, everything, anything north of like, you know, Blore is like another world,

Albert Brier:

yeah, Northern Ontario, like Northern Ontario, much of which is south of central Alberta. But anyway,

Nate Habermeyer, APR:

true story. Anyways. CO generation, really interesting, but we're starting to see more we are.

Albert Brier:

Yeah, it looks like, not only are we seeing a lot of CO generation, but one of the things that I said during the mine Expo episode was that our mining clients are going to start becoming power generation and distribution clients, and that's absolutely happening. A lot of this news is coming out of Zimbabwe. We got mine companies entering the infrastructure market in a mass there, like building, you know, hundreds of megawatts of solar and geothermal specifically to power their minds, but they're able to, you know, contribute to grid power and stability and and all that good stuff.

Nate Habermeyer, APR:

So maybe related to Belt and Road Initiative, yeah, yeah, absolutely a chance when so same question as before, when you're sort of planning, scheduling, risk management, like, what's cogeneration? Like, what's the upside and risk avoidance with cogeneration versus like the other? Like, how would schedulers now that they have the big generation, it's

Albert Brier:

grid stability. So there, especially in places where the grid cannot be relied upon. And, you know, I would say that, you know, we use Zimbabwe as an example, because they surface in this news a lot. So that's definitely a place where you can expect a little bit of grid instability. Like I used to work in Nigeria, same thing, like every building there has what, you know, in our very broad definition would be cogeneration. They've got, you know, generators exhausting directly onto the sidewalks, by the way, super fun. But there's generators, like in every single commercial building there that serve to, you know, supplement grid power and occasionally replace it when the grid power inevitably fails. Sure. So you know, and I would love to say that it's localized to places you know, in faraway lands, but it's not right. The grids are vulnerable everywhere. Whenever you build hardened, robust facilities, you need to think about grid reliability. And, you know, the ability, like, how much uptime do you need? Well, that's, that's your electric too. Like, it can't just be the data equipment, like, for a data center, for example, the electric and the cooling also has to match that same uptime, or something very, very close to it.

Nate Habermeyer, APR:

So it's a lot of, lot of big risk. That's, yeah, big no, so, medium or small risk, huge, huge, big upside.

Albert Brier:

Yeah, so, like, it makes tons, and it's not, like I said, it's not just in Zimbabwe and Nigeria, it's also in places like Texas, right? So Texas, very famously, had some massive grid failures, you know, a few years ago in the wintertime, like, you know, they were getting some unexpected cold winters in a row, right? Two in a row that killed it was many people, and a large like the root cause of this is like the way that Texas manages its grid, which is a whole episode on its own, if we want it to be, it's a bit insane. When it works, it works great, but when it doesn't work, it kills hundreds of people. So, you know, you decide what the trade off is worth. But anyway, the point is that you can't necessarily rely on Texas infrastructure. You're they're getting more hurricanes that are that are bigger all the time. You know most of the people I know who still have family in Texas, you know people who have like retired there. They're retired. Moms and dads are buying generators and. Transfer switches and things. There's like a whole cottage industry around getting those things installed into like retrofitting homes from the 1950s to 70s with these, you know, fairly robust power transfer setups that often include backup batteries and even, you know, don't say it out loud and when you're in Texas, but even some solar, you know it's

Nate Habermeyer, APR:

true, you know, and the infrastructure, but the infrastructure is getting older. Like, there is a lot of older infrastructure that has to be retained, has to be updated. Like, how are they planning for or how would, let's say you're back in Texas, back home, and you're working with, like, some of the companies that you worked for, but, you know, in transmission, in energy generation, you know, how are they planning for these, how would you plan for some of the updates and the maintenance that needs to happen in order to work around Well, the main thing is tragedies

Albert Brier:

it well, yeah, that's really what it boils down to, is like, if the big one comes, what, what do you Do? Like, are you going to be ready? And the answer for most places that exist today is, no, not really, no, we're not ready. So what do we need to do? Is a really good question, and mostly it's, it's, it's about scenario planning and doing some, you know, scenario based risk analysis, like using bow tie studies, for example, to figure out, like, hurricane, right? Big, big fat hurricane hits your refinery. A lot of refineries are right on the coast, because there's, like, marine terminals there for shipping things out. So, you know, the storm surge that happens during a hurricane can be pretty gnarly, and that's actually the thing that causes the most damage on like, a seaside refinery, because you just get those sea water comes and it just floods everything, right? And so you need to start thinking about, what does your mitigation look like? Because it used to be that 1000 year storm would happen every 1000 years. Now it seems to happen every 10 years or so, which means that you got to be prepared to mitigate for circumstances that were borderline inconceivable just a few years ago.

Nate Habermeyer, APR:

Yeah, we talked about climate risk on another episode from last year, which is, yeah, I find it really interesting to explore. But yeah, we did go check out that, if you

Albert Brier:

it is, and I'm basically talking about the same for sure, that one was a lot about adaptation, right? Like, we spent a lot of time talking about climate adaptation. I think that's the thing, right? Like, that's, that's what I'm talking about. Like, that's what a lot of money is being spent on. But anyway, we're

Nate Habermeyer, APR:

getting a little bit let's go. I know it's an interesting No, to get off track, but let's go back to the the prediction. So we were talking about mining, competing with AI for energy was the next one, right? So, yeah, you know, we're on this mining thread, but now we're kind of, like, it's competing for AI, for energy, right? And there's, like, Google, Amazon, like they're all signing contracts to build SMRs. But what is that? Which was

Albert Brier:

another prediction we made, by the way, there'd be more of those, and there have been. But, yeah, we're talking about, you know, I mentioned Texas earlier requests for like, future grid demand quadrupled in 2025 and 73% of that growth was from Ai data centers, right? Yep. So Texas is, you know, as we discussed on that episode, a very comfortable regulatory environment for places looking to build a data center. And that's great. Like, Alberta is another example. Another example of those. But it does going to place a lot of stress on grids, which means that grids are going to become heavily privatized. You know, we have lots of private investment going into grid robustness and grid capacity, you know. And that's one of the, you know, we were talking about China earlier. And the sheer amount, the sheer quantity of capacity that they're adding to their grid, none of that's coming from private sources. Well, I shouldn't say that declaratively, but like the Chinese government, you know, oversees the construction of all this stuff, so they have a bit of a an advantage when it comes to building big things quickly. And in the US, if you, if you want to do that, you're going to have to get some private money involved. And that's exactly what's happening.

Nate Habermeyer, APR:

And I think the planners, the PMOS, like, you know, our audience, the executive sponsors for these mega projects, are thinking about, yeah, different ways of structuring, like financing and power and right? I mean, would they be thinking about a different way of just putting the whole project together and putting it into execution?

Albert Brier:

Well, they would have to, especially if your, if your project is going to be a power hog. I mean, generally speaking, for most construction projects, the workflow is still the same. You have to go to the grid or go to the utility, ask them how much capacity they're going to have in that specific area where you're looking to build, they're going to have to build something new to accommodate your medium to large builds, like a switch gear or something. But there could be a lot more to it than that. You know, there could be, depending on where you are and how big your thing is, like, you know, one of the. Projects that we're working on right now is in Edmonton, Alberta. Edmonton is a medium sized city. If you were to plop a hyperscaler into the middle of downtown Edmonton, there isn't grid capacity to support something like that, so it would have to come with some additional investment in, you know, generation capacity. And the problem with a lot of these projects is their plan on a much shorter horizon than grid capacity in general is. So you know, one of the things we talked about last year wasn't really a prediction, but there was like a run on gas turbines, like natural gas driven turbines, right? Like the pricing and the lead time for those went through the roof because everybody was trying to shore up short term electricity demand driven by data centers. You know, the best way to do that for, you know, economically speaking, is actually solar. But, you know, solar makes poor base load because of the whole Sun is off. You know, no power problems. Sun is off, the sun is off, yeah, somebody reached up, pulled a little cord, and Sun switched off.

Nate Habermeyer, APR:

That's, it makes me think of like a Looney Tunes cartoon, or red and Stimpy or something.

Albert Brier:

That's where I'm at, yes, where I'm at today, mentally, yeah, you could pull

Nate Habermeyer, APR:

you're not at a Beavis and Butthead sort of place. And that's, that's good, yet we're all good.

Albert Brier:

Yeah, fortunately, we're recording this in the morning, you know.

Nate Habermeyer, APR:

All right, so the next prediction, similar, but kind of moving along towards the AI space, but that we talked about autonomous technology risks, and this is, yeah, I actually, I find it's really funny. It's a funny risk. I mean, it

Albert Brier:

is not funny. I'm reading ahead in our research here, and the very next thing is, like, multiple documented collisions, autonomous, and Nate's like, this is pretty funny

Nate Habermeyer, APR:

way to go, Nate, yeah, I just proved that Nate an asshole, and we are lying about Nate dot Dave, I guess. Yeah, obviously there, right? Like we're so what? So the prediction? Let's talk about the prediction. Let's read it, because it's funny. The way that it's phrased, all of these technologies are inherently risky because they're new. Yeah, I think you probably said that. Not me, probably.

Albert Brier:

I mean, there, there is. There's definitely a two sided coin situation here with autonomous vehicles, right? Like, yeah, they are inherently risky. I was just having this conversation with a lawyer friend of mine. The real risk of autonomous vehicles in terms of their broad adoption in public, right? Like on public roads, you know, moving the public around, like Waymo taxis in San Francisco are a good example of this. The real risk there isn't that they're going to be less safe than human drivers. They're not. They're much more safe than human drivers. The real risk totally they are. And by the way, the multiple documented collisions we have here, I haven't, I haven't read deeply into this research, but I'd be willing to bet basically anything that those were caused by the human drivers. Okay, I'd be willing to bet based I'm not gonna take that bet. No, I wouldn't have,

Nate Habermeyer, APR:

but I'll bet you $1 maybe, you know what? Maybe we'll do an episode where we just bet each other money on stuff. Yeah, that would be funny.

Albert Brier:

Quarter bets the episode, just exchanging quarters.

Nate Habermeyer, APR:

Did I tell you that I taught my daughter how to play Texas Hold'em and we were playing like, I'm gonna play her for money? No, just playing tokens. I know a little bit of a tangent there, but yeah, it's, well, it's

Albert Brier:

important to teach them understand risk. Yeah, I was gonna say one of the alternative career paths that is very attainable these days is to go play poker for money. But anyway, back to autonomous haul trucks, the that same strip mining client I was talking about I mentioned on the episode where I first dropped their name. It's Imperial, by the way. You know, no big secrets there. So like Imperial oil is, is, is here in Calgary, and they got out to Imperial, yeah. What's up?

Nate Habermeyer, APR:

Guys on mobile, Yep, yeah, yeah.

Albert Brier:

Not exactly one in the same, but, you know, kind of one of the same. But anyway, so they have a strip mining operation that they would really love, if you didn't call a strip mining operation up in the oil sands that, you know, use a lot of mining technologies, including autonomous haul trucks. And I was part of a very, very small part of the project to get their first autonomous haul trucks in and tested. And they were running a pilot program for a couple of years there where they had, you know, drivers in the trucks, in case the autonomous pilots did something stupid. And now, now, as in, like I had this conversation a couple of weeks ago, all of the haul trucks are autonomous, and their incident rate has gone basically down to zero. Shut up, really, yes, so they don't hit stuff. They're not backing into pipe racks, not backing into loaders.

Nate Habermeyer, APR:

That's because it's just a wide open space and there's no humans. That's why, to your point,

Albert Brier:

and yet, somehow, the human haul drivers were managing to hit stuff anyway. Way, okay, so, like, the incident rate doesn't come down to zero forever.

Nate Habermeyer, APR:

If you hit anything with a haul truck, like you're gonna be okay

Albert Brier:

unless it's another haul truck, right? The other thing is probably dead now, but you know, the point is that, yeah, it is a pretty bad scene when a haul truck rams into something. So the fact that their incident rate is down basically to zero implies that the autonomous part of this equation is not where the issue is. And that kind of brings me back to what I was saying earlier. It's largely a legal concern, right? So one one wonders who is responsible when an autonomous vehicle is involved in a wreck. Like, what if two autonomous vehicles have a wreck with each other? They were both making the best decision available.

Nate Habermeyer, APR:

They can have their AI lawyers Sue and, you know, their AI

Albert Brier:

GPT will write the briefs, you know, the judge will review, yeah. I mean, it does. It does raise some interesting questions. And there, there are damages, right? Like, that's right, there can be, you know, injuries and property damage, Judy and, you know, sure, like this, there's a there's an autonomous Toyota Prius and one side of the courtroom and a person in a neck collar on the other side.

Unknown:

But you know, the reason why

Albert Brier:

I Imperial was able to achieve zero or a near zero incident rate with their autonomous haul vehicles is because they're operating more or less in a vacuum, right? Yeah, like they're fully separated from the public roads. They're fully separated from at this point. Now human haul truck drivers, even like human systems operators, are largely out of the equation. Like the things that that are being, you know, operated, kind of in parallel with the haul trucks are largely remote operated now. So there's a there's a lot less risk just inherent in that system than there was even 10 years ago. So, and that's, I think, you know, the upside, the downside is that there will be collisions and things, and you kind of have to start building a framework for how to deal with that stuff like right now, you know, and we don't have one. You mentioned something to me earlier that I thought was interesting, that the Weibo meltdown

Nate Habermeyer, APR:

in in San Francisco, yep, last December, and

Unknown:

that was because of a power outage, I believe, right, yep, yep. So it all blocked, didn't it blocked

Nate Habermeyer, APR:

trams and whatnot. Like, yeah, the trams in San Francisco and yeah, it was messy.

Albert Brier:

They were, like, there were Waymo taxis piled up at intersections and stuff, just like, waiting for instructions of what they should be doing. And, you know, I don't know the specifics of that, of that situation, but, like, it definitely made the news, right?

Nate Habermeyer, APR:

Let's move on. Albert, let's because we got some other, like, pretty interesting stuff to go over the we do next, yeah, and just to kind of continue on the, I think the AI thread, right? Yeah, it's fertile ground. Look, here, there's so much. We talked about AI adoption and AI sort of the AI reality, like, we are seeing a lot more adoption, yep, but, but it's you know, you could argue, is it better? Is it worse? Like it depends on who you like GPT or not. Yeah, let's read it so that we see what you sounded like on the prediction I

Albert Brier:

made was that 2026, would be the year when we actually started using AI, and that it would get quote more expensive and worse. This, this following the trend of of new technologies, you know, especially new like high technology computer stuff that that comes out, it always seems to start as this unbelievable unlock, this massive value add, and then the price tag shows up, you know, two years later, and it's way higher than you'd expect, and suddenly the services were so that's what I was predicting. I still think that's going to happen. And we have seen, like, there's a lot of uptake, like people are using it more, like the fact that 2026, is the year when everybody starts using AI, definitely, 100% you could argue we overshot that.

Nate Habermeyer, APR:

Like chat GPT is introducing ads, right? Like the people AI is, they're looking for ways to pay for themselves. Essentially, Claude is killing it on the enterprise

Albert Brier:

from they went a to b, and that's going to be their, their big differentiator.

Nate Habermeyer, APR:

I think I was, I was reading an article about the so the developer that that created open claw, or whatever molt bot, or whatever it's called, now, you know, he ended up going to work for open AI, but anthropic actually sent him a cease and desist, which I just thought was a really good move, because if you're in the interspace enterprise space, you're you want to be trusted and you want to be consistent. And so that was a good, good move.

Albert Brier:

Yeah. Yeah, yeah, the knives are coming out in the AI market, for sure, especially when, like, you know, you got Jensen Wong and Sam Altman from Nvidia and open AI respectively, now sparring with each other, you know, out in public, talking about this investment promise that they made the circular investment, which was total nonsense in the first instance. But even so now that it's starting to fall apart in public, everybody's pointing fingers, and these two are pointing them squarely at each other. So I think that this is, this is going to be, you know, I listen to pivot. I know you do too. Scott Galloway compares open AI to Netscape, right? Like the first really excellent browser, also the one that no one uses anymore. I really expect that that's a good comparison that we're gonna see open AI start to sunset in favor of some of the other. Like, I don't you, I've never, I haven't touched chat GPT in probably six months. Like, I've been using Gemini my phone anymore. No indeed, like I the three I use in order, are Claude, Gemini and perplexity. In that order,

Nate Habermeyer, APR:

I use Gemini Claude, and I've started to use Kimi, which is the Chinese app.

Albert Brier:

I'm maybe this is, like, xenophobic of me, but I just, I just don't trust that the Chinese ones. I mean, I just don't, like, I remember when,

Nate Habermeyer, APR:

yeah, I don't, I don't put personal information into, you know, quad. I don't, you know, I think, yeah, trust it. But they, their capability, their context window, is massive, and they are a 10th of the price that would be search. They they are. Kim is killing it. They really are. I'm gonna have actually launched, they actually launched a so not only does Kimmy do Claude code type, you know, work, you know, on your command line interface, they also wrapped, they took open claw and they wrapped it so it's now like sandboxed in the Kimi app so you don't have to set up, you know, your sandbox. Remember all of the security concerns with open claw, and you know, it was just an agent that was running wild. It would take over your whole computer, right? And sending emails. And you're right. Well, Kimmy actually sandboxed itself in their app. And, okay, well, that's sensible. It's cool, like it's for research. It's, it's cool. Cloud code feels sandboxed that way.

Albert Brier:

But every now and then, when it's running in the background, I'll be, I'll be doing something, and then a little banner will appear and say, Claude code is debugging this browser window. Well, I didn't ask it to do that, but

Nate Habermeyer, APR:

okay, Claude co work is really cool, though. If you try it out, it's cool. It's it's essentially, it's like that open claw. It's just a little bit more trustworthy. And yeah, it's good. It's okay. Anyways, we so we digress, but let's continue,

Albert Brier:

not really, though, because we're talking about this adoption stuff, and we pulled some research that shows that only about 12% of the organizations that could be using AI for project management are and of that same population, about 29% feel unprepared to do so. And that's that's a bit of a paradox, because, you know, on the flip side of that coin is in the technology world, like in coding and in high tech development, stuff like, there's already widespread job displacement. That's right, 37% of US companies expect AI to replace a significant portion of their workforce by by late this year. And that, by the way, is another prediction that we made last year. On the same episode, we were talking about massive job displacement, and that wasn't like, wasn't just us, it was everybody right. We everybody saw this coming. Yeah, I mean, the companies that make the AI models are the first in line to talk about how they will replace jobs, right, and replace knowledge workers and things like that. But, like, it's, it's a, it's a huge impact. I mean, I don't remember was it Canva? I think the Canva CEO came out and said that their coders are now just reviewing AI code, like they don't have people writing code so much anymore, like they've they've got a smaller workforce that's just reviewing, you know, AI written code. Amazon shed 14,000 jobs in October. IBM pulled several 100 of their HR workers and replaced them with chat bots. Ignite tech. What you know this guy? I can remember the CEO's name, but he made some some news for himself by replacing 80% of his staff with, well, they weren't really replaced with with AI like llms or what have you. It was more like these were the people that were not on board with the changeover, right?

Nate Habermeyer, APR:

I think there's a lot of you know, if your job is a task, then you should be concerned if your job is a task like that. That's right for AI, if you're, if your job involves thinking, and you know that's you're you're better off. But yeah, if your job is to task. She should be afraid. And we're starting to see a lot of that, lot. It's funny, a lot of those.

Albert Brier:

A big chunk of my early career was spent doing exactly this, like building tools that automated repetitive processes, you know, simple stuff, like in Visual Basic excel programming, like really, really simple automation, stuff that anybody could do. And in fact, no one has to anymore, because you can just get friggin chat GPT or Claude to do it for you. But you know, I got a lot of pushback early on, and it always confused me until I realized that I was putting people's livelihoods at risk by making huge chunks of their job just disappear, right, right? Like, if you have two admins who are doing the job of one person with an automatic tool, then guess what? One of them gets fired, right? Yeah. And usually the one who is the most afraid of getting fired is the one who, you know, pushes back the hardest on on what they're and that probably explains the situation at Ignite tech. I mean, there are definitely people who are racing to the bottom right in terms of quality. And don't mind that the AI is maybe aren't doing as good a job as the people would in those same roles. But you know what, they're a hell of a lot cheaper. So maybe that's enough.

Nate Habermeyer, APR:

Has a What's the I think AI, slop will be one of the words of the year. And you know that annual, oh yeah, like, yes, the, you know the word of the year,

Albert Brier:

slop, yeah, I think you're right. I mean,

Nate Habermeyer, APR:

there's a lot of it. I mean, even now, you look at, I don't know, Pinterest, you look at, oh, it's all Instagram. You look at, you know, even on Tiktok, like it's all, even on YouTube, right? Like it's just,

Albert Brier:

it's Wow. This is actually a nice, clean segue to the next prediction slash topic. You know, we kind of touched on this earlier when we're talking which one is that? Albert, well, I'll tell you. Nate, so we talked about AI needing to build or reactivate nuclear power stations in order to keep up with energy demand. And like I said, we touched on this a little bit when we were talking about mining, but it's a particularly acute problem in AI, because as these AI models grow in capability, they also grow in infrastructure demand. They need a lot of water, a ton of water. They need a lot of power. It's wild. How much resources they already or how many resources they already consume, and it's all just playing catch up right now, isn't? Yeah, that's right. So we're, I mean, you have lots of there were several deals that were signed last year between big tech and nuclear in order to kind of grease the wheels a little bit to smooth that transition. There's more now, right? Like there's already been more announced, some in January, one signed by meta. That's the big one. Six gigawatts of a dedicated power being diverted from existing plants and future SMRs. And these are all there's a mix of old and new nuclear technologies in there. Google signed, I think it was just an MOA for off take from fusion companies. Those fusion power doesn't really exist yet. But you know, all these companies are looking towards the 2030s

Nate Habermeyer, APR:

for when it will beyond, yeah, 100%

Albert Brier:

Yep, so, and this is, it's, sorry, go ahead.

Nate Habermeyer, APR:

It's gonna be, it's it's gonna be, it's a race to catch up. You know, at this point, I gotta imagine that the project managers at some of these tech companies that are trying to figure out, you know, the roadmap for, you know, new products and services. Now it's just when is when am I going to get the power? Yeah, yeah, to run right. That's the biggest dependency,

Albert Brier:

it is. And we quoted some stats around the growth of data center power demand being something like 20 to 30% compounding annually. And it's turned out like the 25 to 26 growth was a little lower than that. It was around 14% and that's the new projection out through 2030, which is, you know, a good thing, but that's still 100 gigawatts of new capacity. That's $3 trillion in investment required by 20 $33, trillion now that's a lot lower number than it used to be, right? But, you know, because that's like the market cap of name a tech company, which is insane, that there's multiple that are over a trillion, but they are, yeah, you know,

Nate Habermeyer, APR:

I feel it's like only yesterday that who broke, who broke the trillion? Was it Microsoft first?

Albert Brier:

I don't think. I think it was Nvidia. I think it was Nvidia broke a trillion first. We can get pretty easily fact checked on that one. But the point is that there's now multiple that are over a trillion.

Nate Habermeyer, APR:

There is a listener survey. If you want to fact check us, you can go to our show. That's right, you can fill out that listener, please,

Albert Brier:

go tell us how wrong. Please, please abuse us. Tell us that we did a bad job over it. Nate, not Dave, do.

Nate Habermeyer, APR:

You're right, 14% 100 gigawatts of new capacity, $3 trillion how are you like what's going through the mind of an executive at a power company? Are they excited by this, or they is this? I have to

Albert Brier:

well, on the one hand, on the one hand, well, there's certainly going, you know, they're making a lot of phone calls. They're showing up at a lot of private islands to talk about how amazing their upcoming power technologies. Yeah, different private islands, the kind where you play golf and do business, not the kind where anything else might happen. But the idea here is that you're gonna have to a lot of these deals are being made. They're not being made, you know, through an RFP process, they're being made because people know each other, right? Yeah, so there's a lot of glad handing going on right now. I would bet the other thing that's, you know, on the one hand, it's dollar signs, right? Like it's big, big money to get involved with one of these, you know, trade deal, theoretically, right? Because all this capacity has yet to be built. Which brings me to the second thing, like, that's a lot of, that's a lot of projects, right? A lot of projects that suddenly just need to appear out of thin air and get built in the next five, six years. Like, that's, that's a lot, you know, we don't do great at capacity enhancements on the North American continent. Like, we're not amazing at it. You know, one of the things that that we we mentioned in our nuclear episode, one of several nuclear episodes last year, was that capacity enhancements and life cycle extensions were a more viable path to a long term nuclear future than the new builds were at least as of that moment, and that there continued to be, like, you know, 10s of billions of dollars invested in Canada alone in renewing old nuclear power stations to make them good for the next 30 to 50 years. So the

Nate Habermeyer, APR:

executives are happy, right? Because they're glad handing they're shaking hands. They're going golfing. But it's the, you know, wake up in the middle of the night with a cold sweat, stress that's happening at the program,

Albert Brier:

at the project executive level, yeah, like, right, all the because those people are the ones who actually have to deliver on those promises, right? Like, CEO gets to play golf, have a nice meal and, you know, sign a deal that has a $400 million you know, bottom line, and that's all wonderful, but the program managers and project executives that are responsible for delivering that now have, you know, a massive peak of effort to climb, you know, right? Like, it's going to be, it's already, it's hardly hard enough, right? Like, I've worked on enough of these refurbs, enough new builds to know that, like, it's pretty hard under normal circumstances when you aren't up against the clock, and we are very much against the clock right now.

Nate Habermeyer, APR:

I you know what this is. This is an interesting one, but I'm going to read this prediction because I bet this was what I said. We're going to have to as a species run an enormous amount of new fiber optic cable in the next 10 years to keep up with all this demand for bandwidth. Actually, that was a you, that was not me, because I would not have called the human race species.

Unknown:

Yeah, there are two as a species quotes in this document.

Nate Habermeyer, APR:

It's true. This is an interesting one. Like, so fiber optics, right? Like, who you know, of course, like, you're gonna need to move data faster. And, like, just last week, I think I read an article about Corning stock is just through the roof because they're racing to meet demand for a new fiber optic cable.

Albert Brier:

So I heard another story along these lines that I thought you might find interesting, pray tell, well, so, you know, Toto, right? Toto, toilet seat, Toto.

Nate Habermeyer, APR:

Yeah, of course I did a Renault and I put totos in my house.

Albert Brier:

Yeah, of course you did, because of the best, right? Like, I've got a, I've got a toto washlet sitting, you know, not 10 feet from me, right there, fancy Japanese toilet seat. So toto has some pretty significant ceramics manufacturing capability, as you might guess, and they supply the brackets that, like memory chips and like integrated, you know, silicon are made in right? Yeah, like, super high test ceramics are important for a process called cryogenic etching, which is when they use, like, ultra cold temperatures to, like, build layers, like, at the molecular level into memory chips, which is how we're doing it now, okay, like, Toto has had like, massive growth in their stock price as a result of, you know, getting part of because, like, memory cost is crazy right now. Like, sure, like, a buddy of mine surfaced a screenshot to me showing that, like, buying 64 gigabytes of RAM is more expensive right now than buying an AR 15, right? So that does not seem like it's well calibrated, frankly, but like, that means that there's a lot. Lot of market cap making memory.

Nate Habermeyer, APR:

I want, I want, you know, lots of gigabytes of RAM. But I don't, I don't, yeah, that's a really

Albert Brier:

interesting but you don't want it

Nate Habermeyer, APR:

more than you and I having born and born and raised Americans. And we've talked about this before. Americans have this thing where we will measure anything other than, like, an appropriate measure, right? It's like, oh, it's three football fields. Or like, Oh, it's just as expensive, or more expensive than ar 50. Like, we will, yeah, right. We always will measure

Unknown:

it. But I still think that's like, 65

Nate Habermeyer, APR:

tankers long. I don't know

Albert Brier:

how many tractor trailer trucks, 4.35

Nate Habermeyer, APR:

Eiffel Towers tall. Yeah. Well, this is interesting because we're talking about fiber optics and ceramics that are holding, like,

Albert Brier:

like trying to make, is that, like, the this, this issue of, like, needing more ceramic capacity to do more computer stuff, like, it's causing toilet manufacturers to have a banner year. That's where I'm going with this. That's kind of nuts. Okay, so anyway,

Nate Habermeyer, APR:

yeah, yeah, let's move on. So one of our best, one of my favorite, episodes from last year, and some of the predictions like you went to the AAC e in San Diego, right? Yep, yep. And you recorded interviews on the floor. Going again this year. Going to do the same thing, yeah, but not in San Diego, no. And you're actually going to be presenting at AACS.

Albert Brier:

Fingers crossed. Nate, fingers crossed, because, you know, we just submitted so me and my co authors, whom I will name drop. So it's myself, Roger Bradfield, also from do kanish And Rachel Fleming from MVP consulting, we co authored a paper on program risk management, which, you know, mark your calendars. It'll be presented at the Vegas show this year. Unless it isn't right, it still has to clear peer review and go through an editing round and stuff like that. We submitted it about three weeks ago, so we're waiting to hear back on the first round of edits. But we were not the first. So everybody asked for extensions, right? Because the paper submission deadline, just like bookended Christmas, like they accepted the submissions on, you know, in like September, October, and then they were accepting them and at the end of January. And it's like, what do you guys think is happening between October and January, just like brainstorm a few things.

Nate Habermeyer, APR:

Yeah, not much, not much, other

Albert Brier:

than Halloween, Thanksgiving, Christmas. And that's just to name drop the Judeo Christian ones I'm familiar with. There's plenty of other stuff going on. But anyway, the point is, it was a busy time, and lots of people didn't submit on time. So I'm not expecting to get comments back for another week or two at minimum, fair, but the

Nate Habermeyer, APR:

comments that you got when you enter, well, hold on, let's go back to the actual conversations you had. Sure, yes, Laura, because those are really interesting. So one, we did an episode where we talked about those, and then we you did, we did another episode. You just went through and played the recordings, which was really cool. So you can go check those out on our on our podcast,

Albert Brier:

and we will do that again. But like the the main standout concept there, we talked about the California High Speed Rail. Locks were in California, so why not high speed rail? You know, one of the things I mentioned on that episode is that any rail project of that magnitude is going to be a mega project like it just is, like there's, there's too much money required to do large infrastructure development for it not to be a mega project. And nine of 10 mega projects go over on cost. So we talked a lot about about that at the the show, and people had some interesting takes on the on the high speed rail project, specifically. But it must be said that, you know, a pretty big chunk of the of the the the segment that was approved, like a little tiny, you know, 120 mile segment in rural California, rural esque California, a big chunk of it's already been built. That's right, built, yeah, like, there's no track there yet, but, like, they've built a bunch of it. There's, you know, right here in Calgary, the green line, which has been a perennial failure, like, something that just keeps not happening, like, that's starting to get built too. So maybe we need to put our flag down and say that 2026, is the year of trains.

Nate Habermeyer, APR:

Well, you know, here in Toronto, the Eglinton Crosstown is a tram, and it opened, finally, a billion dollars over budget.

Albert Brier:

Yeah, that one was a feature of a previous episode from this year. I think

Nate Habermeyer, APR:

it was, but it's great, and it's nice to have more transit. I mean, if you're gonna be a, you know, world class city, you need to have the transit. But when you were at AAC E, you were talking about the high speed rail, and what did people say? What did people have to tell

Albert Brier:

Well, there were, I would say it was, it was a pretty solid majority that were not optimistic about it being built. I mean. Yeah, go figure. Yeah. I mean, taking history as a guide, obviously, things have not gone great for that project. But if you just like, cast a wider net, like rail in North America is just fraught with issues, like, absolutely fraught, like when, when the car took over, you know the American consciousness in the 1940s and 50s, like investment in commuter rail just like, stopped, right? Yeah, I mean, except in some of the places that already had it, right. Like, you know, New York is has a famously both good and bad trade network, right, which is getting a lot better as I understand it. But the reason why New York City has such an extensive commuter rail network is because it has had one for over 100 years. Yeah, right. Like, the first commuter rail in New York was a pneumatic rail system from like, the 19th century, you know, so and some of those tunnels, I think, are still used for the for the subways. I might be wrong about that, but the point is that if you didn't get there early, you don't get there at all. So having right, you know, this massive investment in building new rail is like, it's promising, but you know, once again, the people doing it best are the Chinese, and it's not like we're inviting a bunch of Chinese contractors to come here and build our rail for us. Like, no, no, we did, but we're not, you know, yeah, yeah, they were invited, yeah.

Nate Habermeyer, APR:

I think as of August 2025, 59% of that segment that you referred to is complete now, yeah, yeah, right, right. So it's progressing, which is good for that rural area of California that's connected

Albert Brier:

now, yeah, I mean, we're only dealing with 35% cost overruns on that section, so I guess, compared to some of the other ones, we've got, like the channel tunnels at 80% you know, if you include financing, it's at 100 140% you know, the Big Dig was 220% over. Percent over the Opera House. You know, we talked about that on our last episode. But like all these, like major infrastructure projects, that opera house isn't really infrastructure. But the point is that they all tend to run massively over. So 35% on this segment is pretty good. But I do wonder where that, you know, is measuring from. It's certainly not measuring from the initial high speed rail estimates from like, you know, 10 plus years ago. I mean, like those numbers have, you know, the planning eclipsed the full construction cost of those numbers years ago. So there's no chance we're talking about 35% versus that must be against a recent baseline.

Nate Habermeyer, APR:

What's your one piece of advice for that project executive as they look to some of the new technologies in 2026 and how it might enhance some of their project delivery.

Albert Brier:

Well, I'm going to touch base to the one that I always fall back on, which is AI. I mean, we dropped a stat earlier that like you know, less than 30% of companies feel confident they have a path forward for AI. Meanwhile, there's, you know, 10s of 1000s of jobs being shed each year, and that trend will likely continue as AI uptake increases. So that's an odd convergence of we're already letting 10s of 1000s of people go, and we don't know what to do with their replacement tools, right? Like, it's time to figure that out, right? Yeah, we talked about AI enabled tooling, and some of because AI tends to be a conversation about large language models. And that's not the only AI technology that's maturing fast right now, like machine learning, more generally, from like a pattern recognition and data analysis perspective, like those tools are getting democratized too, but they're, you know, llms are sucking all the air out of the room so it that conversation hasn't really matured yet. It's time for it to like, I think it's time to start thinking about pattern recognition machine learning as your next big unlock for taking your historical data and turning it into a planning tool. You know, like right now, a lot of companies, including some of our clients, have whole departments of estimators and project managers who are just wrangling historical data. Well, as soon as that data gets wrangled, it's going to get fed to an AI as a training model, and we're going to start seeing AI planned projects, I think, this year, where AIS have crawled historical data and pulled together like schedule estimates and budgets, sure at a high level, but enough for a screening that'll probably be a heck of a lot more accurate. And oddly, you know, we talked about strategic misrepresentation and the cognitive bias against the big number on a recent episode. I think that if executives are seeing the number come out of a computer system, you know that is theoretically, not theoretically. It isn't a person, right, like you can't hate it, you know, it's just. It's just numbers, and I think executive is going to be more likely to trust that than they will be to trust the many, many schedulers and estimators who've been telling them for years that their numbers are too small, right? You might actually see executives start to take that seriously as soon as AI says it instead of say, Me, for example. So interesting,

Nate Habermeyer, APR:

so interesting. Thanks. Albert,

Albert Brier:

yeah, thank you, buddy. This was a fun one. I enjoy reading my own predictions.

Nate Habermeyer, APR:

Think, yeah, the transcriptions will get some funnier, because I think most of them were yours, not mine. Mine again would have been all over the map, but yeah, thanks, man.

Albert Brier:

Predictions though, yeah, the next one will be even more fun.

Nate Habermeyer, APR:

I'm gonna talk to 40 more Nate's, and we're all going to, oh yeah, confirm that we all get called Dave. Anyways, thanks a lot, Albert, thanks buddy. All right, talk to y'all later.

Albert Brier:

Bye. Hey everybody. It's Albert here. Thanks for tuning in to the risky planner podcast. We hope today's conversation was informative and, above all else, inspires you to excellence in what you do. If you liked today's episode, don't forget to rate, subscribe and leave a review. It helps us reach more listeners just like you. I'd also like to thank Thompson Igbo Igbo for letting us use his excellent music on our show. If you like what you hear. Check him out@egbommusic.com that's E, G, B, O music.com talk to you later. You.

Podcasts we love

Check out these other fine podcasts recommended by us, not an algorithm.