SPEAKER_02: 0:00

So we are still in a situation where there is a productivity crisis. The difference being is that these AI tools will mask a lot of the productivity issues.

SPEAKER_00: 0:16

Hi, and welcome to another episode of Stell Me This Podcast. This week we're joined by Tim Chen, who is one of the co-founders of Untapped Energy. We covered a lot of ground in this episode talking about everything from AI literacy, AI fluency, and the importance of data as a foundation for getting your AI strategy right. I can't wait for you to have a lesson. Welcome, Tim. We're super excited to have you on another episode of Sell Me This Podcast. We are going to dive right into the conversation today. And why don't you kick things off by just introducing yourself and tell us a little bit about who you are?

SPEAKER_02: 0:48

Absolutely. First off, Keith, thanks for having me here. I am so delighted. My name is Tim Chan and uh born in Toronto, uh, but moved to Calgary in 2005. So for the most part, I do consider myself a Calgarian. Uh in the daytime, I work as an assistant professor at the Bissett School of Business at Mount Royal University. Uh, I'm also a co-founder and director of Untapped Energy.

SPEAKER_00: 1:13

Oh my goodness. Okay, so there's a lot to unpack there already. And I know that we're hot off the heels of a really successful Untapped Energy Education Summit, which I think we can we can dive into. But both of those things um seem complementary but but separate. How did you find yourself um doing the work of both of those things?

SPEAKER_02: 1:31

Actually, it was all by accident. Uh so maybe I'll share a little bit of the origin story of Untapped Energy. Uh back in 2018, uh, which really felt like such a long time ago, uh, there was a lot of chatter and excitement about this new topic, big data, that somehow big data would be sort of the mechanism by which organizations would be able to create more value, particularly in the oil and gas sector where I was working at the time. And at the time, there was also a lot of growing concern, as is typically seen whenever there is any disruption when it comes to business processes or technology. And so there were a number of oil and gas professionals, uh, primarily geologists, geophysicists, petroleum engineers, who are starting to wrestle with this idea of, well, if all of a sudden the organizations were going to lean heavy into this big data concept, did they have the skills and the competencies to continue to thrive in this particular sector? And as a result of that, many of them became concerned. And so that's really how untapped energy started. So there were a number of geologists and geophysicists, petroleum engineers in oil and gas that got together and thought that wouldn't it be great if we could create a community of professionals in oil and gas and to organize events that could showcase what it would be like to collaborate, to come together, and to solve problems using data. And so the idea was born to create the very first oil and gas data fund. Now, they were, I guess, going through their networks to try to see who they can get representation from some of the large oil and gas companies. And then they came to my company. Now, I was third on the list. Uh, they had actually two other data scientists that they had wanted to get involved. Uh, as fate would have it, those two were not able to attend the meeting. So they're like, okay, fine, we know this accountant from this organization, let's give them a ring. Uh and turns out I attended, and the very first question I asked was, Well, what is a data thon? Well, I would learn that it is similar to a hackathon, except you gather people, you provide them with the data sets and with some data tools, uh, give them uh three days uh over the course of the weekend to then try to solve some problems using these particular tools and techniques.

SPEAKER_00: 3:50

Aaron Powell Very interesting. And so, yeah, I think there's a lot of people that are very familiar with the hackathon. And so the data thon, you almost start the other direction, which you say, okay, we have this large set of um either structured, I'm guessing, or unstructured data, and and kind of what what can you find out of it? Like how how what does it actually feel like to be part of that kind of experience?

SPEAKER_02: 4:10

Aaron Powell Well, it's amazing. Uh and first off, when you were talking about whether the data was structured or unstructured, uh I would call that it was all messy data.

SPEAKER_00: 4:18

Yeah, it's it wasn't tied to a there's no bow across it.

SPEAKER_02: 4:22

We wanted to take a look at some of the biggest challenges facing the oil and gas sector at that time. Um there were a lot of headwinds facing the industry. You know, you had uh takeaway constraints, so there just weren't enough pipelines to move some of this crude product. Um there was also uh government regulations that weren't very favorable to the particular industry, and it was an industry that was very sensitive to market fluctuations for commodity pricing. So all these things were creating challenges for a thriving sector. And so the data fund essentially went out to look for publicly available data sets related to oil and gas. Now, little did we know that that was such a wide net to cast, and what we eventually got back was just messy, messy data that required a lot of cleansing of the data, a lot of structuring, uh, and ultimately it led to uh where there were five specific questions that we uh were trying to solve. Uh everything from corporate social responsibility to the repurposing of existing oil and gas assets uh to clean water usage.

SPEAKER_00: 5:31

So, as the individual that uh, you know, either by through the lucky straw or the short straw, as you attended that first meeting, did you have a lot of knowledge in big data at that time, or was this something that was more of a passion project that you were just excited to be a part of?

SPEAKER_02: 5:45

I just naively stepped into something that I thought was cool to be part of. It turns out that I was surrounded by a number of data scientists, uh, mathematicians, computer scientists, those that you would typically think would be in this space. And so I had the greatest form of imposter syndrome that entire weekend. But I pushed on, just thinking this would be great to build a community around just wanting to solve complex problems. Well, halfway through the event, as I was having some serious doubts about my involvement with this event, someone came up to me, my friend Sheldon, and he said, Tim, you need to put aside this doubt that you have. You're an accountant, which means that every month you're going into the organization's enterprise resource planning system, you're extracting the data. Uh it turns out that the data that you pull out, you really can't use it, so now you're having to transform that particular data. And then after that, you're doing some analysis, drawing some insights, and then you're loading that data back into some sort of management system. He said, Tim, that's ETL. As an accountant, for the last decade, you've been doing that every single month. And it was that aha moment that really set aside all that worry, and I'm like, actually, I do belong here.

SPEAKER_00: 7:02

I find it really interesting. We we talked to a lot of different professionals that aren't in technology and they believe that technology is somewhere else over here on the off to the horizon. It's something that supports what they're doing. But there's so much technology practice that actually comes into play. Um in that exact I love that example that you shared where you know you were you were doing that work already, you were you were building those processes, you were really embodying the exactly what that group needed, but you you didn't even have the you just didn't have the right label for it.

SPEAKER_02: 7:33

Absolutely, Keith. And sometimes maybe that's all that's needed in order to make a difference. You just need to have a little bit of curiosity that allows you to just extend yourself beyond your zone of comfort uh and always be willing to learn from others. So it there is a little bit of that humility uh that really does open up uh different doors.

SPEAKER_00: 7:52

So speaking of doors opening up, how did you then transition from this really inspiring weekend data-thon to the founding of Untapped Energy?

SPEAKER_02: 8:02

Well, after the weekend was concluded, uh we sat back and we're like, wow, this is so amazing. Like over 125 people just showed up out of the blue, like voluntarily. Nobody was paid to be there. This was when the Bull Valley College opened up their downtown campus. So they graciously offered that space to us. And we thought, we need to do this again, but we can't wait until the next year's data fund before bringing back together such a compassionate, impassioned community of people that just want to do good and solve problems using data. So that's when we said, well, why don't we start meeting on a more frequent basis? And that was the start of our monthly meetups. And so since 2018, Untapped Energy has been meeting on a monthly basis where we would invite speakers to come in to share a little bit about their challenges in using big data, but also to maybe highlight how it is that they've overcome some of these problems. And we have found that that type of gathering has been so inspirational because people could be wrestling with the exact same problems at that moment, hearing how someone has solved it for them.

SPEAKER_00: 9:09

And who are the types of people that you're finding at these meetups? Is it is it big data experts where they're they're diving into all of the data, or who would who would be attending these types of meetups?

SPEAKER_02: 9:21

Well, initially we thought it was that type of persona, right? The person who's spent many years doing computer science or into machine learning. And certainly uh we had many of those participants show up to these meetups, but then we started to see others in business showing up, the accountants, the lawyers, those that are in HR. Because it turns out that in this day and age, everybody is dealing with data in one form or another. If you send an email, you're dealing with data. If you are having to update a spreadsheet, you're dealing with data. And it also turns out that a lot of these users of data wrestle with a lot of the same challenges and are motivated and curious into how it is that they can either make their own workflows a little bit more efficient or to solve some of the more complex problems.

SPEAKER_00: 10:09

Very interesting. And so when you were first starting these meetups back in 2018, was AI part of the conversation or was it very much around the data itself?

SPEAKER_02: 10:18

The wonderful thing about the city of Calgary is that there is a very vibrant grassroots group and movement where uh if there's any type of interest, uh, there's probably a meetup group for that. So very quickly we started to see that for a data community, they were also part of coding communities, or they were also part of machine learning communities, and they were also part of an artificial intelligence community. And the generous nature of this particular city is such that everybody is always looking for ways to collaborate, to partner, and to help each other up. It's this uh idea of you know, if you can lift the entire tide, then everyone gets the benefit. And so, yes, even in 2018, there was, I'd say, kind of a smaller movement around artificial intelligence. It wasn't yet at the stage where it had captured a lot of culture's imagination. But still there were those that knew all about neural networks and these transformers, all of these things that you know maybe some five years later we would eventually hear more about.

SPEAKER_00: 11:19

Well, on the AI topic as well, like I I believe that's something that a lot of people don't fully understand is that it isn't just the release of Chat GPT that has kicked everything off. This is a conversation that's been happening not for for years, but for decades. Um and it might have had different labels, it might have had different um pseudonyms, but uh but this is a conversation that has been happening for a very, very long time, and data is an incredible part of that, and I think people are starting to see that more and more now, too. So, how has the the conversation around data and untapped energy changed with the momentum around AI right now?

SPEAKER_02: 11:59

Yeah, certainly I'd say that machine learning, uh deep learning, and then artificial intelligence has always been a feature of a data community. And I would say that certainly with the release of ChatGPT in November of 2022, where now that seemed to be kind of a groundswell that has just exploded and now everybody is aware of that. But what it all has also done is reminded to our community the importance of data and data literacy and data fluency. Because all of these algorithms that are running a lot of these generative AI tools that we've now become familiar with, it all starts with data. We keep hearing about how all of these large language models they need to be trained on data. Where do you get the data? And so for someone who is well fluent in data, uh, they have now the skills and capabilities of knowing where the data is, uh, knowing how to create the necessary pipelines to extract that data in an efficient manner, and then doing something with that data so that you can put it in a way that's a little bit more structured, that you can actually start running some work on that data.

SPEAKER_00: 13:07

And so I don't want to get too technical today, but I think you have a really interesting perspective around the importance of that foundation. And so if if you're talking to a listener, if you're talking to a business owner that knows that they have to walk down this path, they they know that they need AI, they know they need all of these things that everyone's talking about, but haven't the faintest clue where to start, but they also hear this narrative that your data needs to be in order, your house needs to be in order. What are some of the things that they should be considering or contemplating to start walking down that path? Or is there any low-hanging fruit that exists for them as someone that might not be as intimate as you are?

SPEAKER_02: 13:46

So there have been a number of AI pilot projects that have been completed at this point now. Uh unfortunately, many of them have not moved on to where they're now out of the pilot phase because of some of the results that were observed. And one of the main reasons why these particular pilot projects are not achieving the expected or desired outcomes is down to data. Um the data is just not in a good enough state, which enables a lot of these tools to be able to really leverage what's in the data that can create some amazing insights, which would then enable the organization to make some quick decisions or to optimize some of its processes. And so for a leader of an organization, knowing that this has been the trend, that data has been one of the downfalls of successful AI implementation, what they can do now is really be proactive of ensuring that even before any type of AI initiative is being proposed, that the data is in order. Now, one of the challenges for a lot of organizations is that they've inherited data from various legacy systems or legacy approaches. All done while, let's be honest, people didn't know what data to keep. People didn't know how it is that they should label their data or store it in such a manner that followed some sort of structure. And so you end up having uh these piles of random Lego bricks all over the organization that you know that at some point you might be able to draw some sort of insight out of those Lego bricks, but it's in such a mess. So by the time an AI project comes along, uh, you know, they have to do that type of uh put a lot of effort into that, and that can sometimes eat up a lot of the actual time and effort of an AI project.

SPEAKER_00: 15:40

And then so if I was to try and unpack that a little bit as as someone that might not have that same technical depth, are there things that I can do to start to put those Lego pieces in a way where they make sense?

SPEAKER_02: 15:52

Well, I guess the first is just to know where all the Lego bricks are. So even if there's a way to come up with some sort of mapping or documentation of where the data resides. They're typically in databases, uh, so that shouldn't be too hard of an exercise to complete. But then once you've got a uh a good sense of where the data is, then it is coming up with a particular framework of how it is that the data needs to be categorized. Let's put all of the red bricks together, let's put all of the green bricks together. How about the ones that are maybe two by two and the ones that are you know two by eight? So having some sort of structured approach uh as a framework is really the beginning of what's known as data governance, where it is that you have guidances on how it is that your data uh should be stored, not unlike how organizations treat some of their merchandising inventory or some of their um hydrocarbon inventory in the case of an oil and gas organization.

SPEAKER_00: 16:51

So that makes sense to me. So kind of what you're suggesting is saying we need to almost define the rules to play by for data, understanding how it traverses our organization, where it lives, what street address it's on, and and really kind of creating that infrastructure so that we don't have to make an individual decision each time.

SPEAKER_02: 17:08

Absolutely right. Because once you know where your data is and in what state it is, that's when you can start actually using it.

SPEAKER_00: 17:17

Can I ask a little bit of a candid question? Sure. Like I believe that everyone knows that this is super important, or at least at least maybe the people that we talk to in the circles that we talk to. But this seems to be a part where everyone falls down, and maybe it's because it's not the shiny object and the exciting work potentially. But but why don't more organizations invest in this foundation? Because I I feel like this is exactly why a lot of them struggle with the actual ROI on their projects.

SPEAKER_02: 17:46

Yeah, it's such a great observation. Um, and I think it comes down to courage. It's the same type of courage that causes someone to go into the basement of an old building not knowing what's there, but knowing that there might be a swamp of mess, messiness. And so uh a lot of leaders who have the motivation may not have the full courage or support to wander into the depths and the dungeons of where all this data is. Um so part of data fluency is arming yourself with enough tools and a mindset so that you can then put on the right equipment, get the right lighting gear, to then wander into this dark dungeon uh to see exactly what the state of your data is in. And oftentimes, once that happens, that's really then kind of the tipping point where then organizations feel more confident about how it is that they can pull the data, uh, be able to uh make it more current and more relevant and more useful.

SPEAKER_00: 18:52

So you said something very interesting there, which was data fluency. So, how would you describe data fluency in today's business environment?

SPEAKER_02: 19:00

Yeah, oftentimes we hear about data literacy uh as an important upskilling capability that employees should have, workers should have. Uh I call it data fluency because just like if you and I were to go to a country where we did not speak the language, yeah, you can jump on Duolingual and it'll give you a number of uh activities, exercises to help with your literacy. So you may learn the characters or the alphabets, you may learn some of the rules of grammar in order to put those things together. But that doesn't make you fluent because the moment that you land in that country, you'll realize very quickly that you cannot communicate with someone who has that fluency. And so there is a big difference there. There's certainly lots of content information out there to help people on their data literacy journey. But fluency comes from mastering a lot of what you've learned in literacy and being able to put it all together so that you can communicate in a very compelling and inspirational way. So, this is where data literacy is sort of the aspirational point for someone on their data journey. It will start with data literacy, uh, learning about databases, maybe even learning some of the coding languages that helps you to clean and organize your data. But really, where the value comes from on that journey is being data fluent, where you can now interpret the data, pull out meaningful insights, and be able to communicate that in such a way to key decision makers within the organization.

SPEAKER_00: 20:36

So does this work go away with the increase in AI tools? Um, you know, a question that I've received before is can I just dump all my d data into the AI and I'm doing some air quotes for anyone that's just listening, and then it'll just clean it up for me. Is that how it works?

SPEAKER_02: 20:55

Well, certainly, as we're seeing now, there are a lot of AI tools that can help with some of the monotonous and repetitive tasks in a particular workflow. Um we've seen this before in maybe prior iterations, before it has this AI label in the form of automation or RPA, robotic process automation. Uh AI can certainly uh provide some of that lifting of the tedious and monotonous tasks, which you can find in a lot of the data cleaning up uh aspects and processes. But still, in order to get to the point where the insights that are coming from that data analytics work, uh, you still need to have that human intuition, that critical thinking that is needed. So there's recently been a report that uh came out that said that 40% of employees that were surveyed are now indicating that they're seeing something called AI work slot showing up. This is now work that is clearly generated by AI that looks and sounds like it's good, but what it's what it's causing is now human workers have to now stop what they're doing and review and inspect this other work to determine whether or not it's actually accurate. And so I'm not sure if a lot of people would have predicted the prevalence of work slop now showing up and how that's actually working counter to some of these productivity aspirations for the organizations. So, similar to being able to determine what is good work and bad work, uh, those that have the data and fluency, data fluency would have the competencies to be able to determine what is good data analytics work and what is not.

SPEAKER_00: 22:45

That makes a lot of sense. And you know, uh to build on that, I was out for a coffee with someone the other day, and we were talking about one of the vibe coding tools, one of the new coding tools, and it's incredible what you can do. But under having a foundation of actually how platforms work, you can take those actual vibe coding tools and take them way further. And so, an example, I you know, I had a small use case for Deliver Digital, I built something out, and I was able to make something a fairly functional MVP. In the same amount of time, the gentleman that I was talking to, he runs a software development company. He built something that you could probably put on uh, you know, start going for funding rounds tomorrow. Like it was a very functional SaaS application that blew mine out of the water. And just because he knew exactly the right questions to ask, the framing, the context, data structure, and having that foundational knowledge is is a must.

SPEAKER_02: 23:41

Absolutely. So this is not at all discounting the fact that uh for someone to have valuable AI skills, they should learn prompt engineering, they should learn context engineering, they should learn vibe coding. But I believe where humans will continue to still have the advantage is how do you act as a leader in AI? So what does it mean to implement AI responsibly and ethically? Uh how does these AI tools impact change management, which has always been something that uh leaders have always been wrestling with.

SPEAKER_00: 24:18

Definitely, and I think to your earlier point as well, how are you managing these these AI either agents or workflows? Because input and output are are two very different things. And if you put in a crappy prompt, or if you put in a crappy request, you're gonna get crappy information back. Yeah. And if you continue passing that down the line, eventually you're just gonna have a pile of garbage.

SPEAKER_02: 24:42

Absolutely. So I I believe that there is a crisis that we're facing right now with AI, and it is called a productivity crisis. And it's not a crisis that we haven't seen before. So actually, Keith, if you think about it, most of why organizations go through a digital transformation is because someone has identified a productivity crisis. We need to use our data better in order to generate more value without incurring more costs. So we are still in a situation where there is a productivity crisis. The difference being is that these AI tools will mask a lot of the productivity issues. So a lot of incentivization structures, or even the way that organizations are structured right now, is more on input. How much can I demonstrate that I'm putting into the process? And so this leads to then the creation of miles and miles of PowerPoint slides or reports that you can create entire cottage industries around just doing stuff or appearing to do stuff. And don't get me wrong, these AI tools are great at making you look great when you're just doing stuff. But the crisis is this is that organizations who aren't focused on the output are going to be caught in a bit of a loop where now workers would become better at using these tools that generate slob or create things that demonstrate that they're productive on an input perspective, but it does nothing to show that they're actually successful on the output perspective. And so I think the moment that we can maybe address that crisis through AI fluency, through appropriate leadership mindsets and sensibilities, this is when the needle will start moving on actually seeing the true value of AI show up in organizations.

SPEAKER_00: 26:31

So how much do you feel like leaders need to shift their perspective then in terms of even what their goalposts are? Because you know, I remember when I was early on in my career, um, you know, starting to manage people, I was told over and over again, you know, you can't throw more people at a bad process. And occasionally you can, you know, you can move the needle forward, um, but are you moving the needle forward in the wrong direction in an effective way and masking to your point the problem by just having more people? And AI just seems like a cheat code to say, okay, well, we can times that by infinity now, because the people aren't the constraint, and so we can just kind of continually double down, show progress in a way, but it might, to your point, not be actually leading us anywhere.

SPEAKER_02: 27:16

You're absolutely right. So I'm suggesting it's not even shifting the goalposts. We need a whole new game altogether. We need a whole new goal, and this is where AI fluency will help to um highlight that so that you're not just using these tools to try to exasperate an existing crisis.

SPEAKER_00: 27:37

And so, how does someone even start to wrap their head around that? Like if you know, if you're if you're talking to me and I own a you know$30 million a year construction company, and you say, okay, great news, uh, we need to change the goalposts completely. Um we need how do I how do I even start to contemplate that? Like what are the things that I need to be doing to get myself into a headspace to be able to see how I can change myself, my business, my goals, etc.

SPEAKER_02: 28:02

And this is not at all to suggest that you get out of that business altogether. You're still in construction, and so you know that at the end of the day, the key thing is that you're gonna be building things that are of value to whomever your customers or your clients are, and you need to do it in a safe manner, right, so that the people that are supporting you get to go home every night. AI fluency just uh causes you to think about the problem in a different way. Whereas traditionally we are so used to you have to do things in a very linear fashion. You've got to do this before the next task can be completed or started, and that needs to be done before the next task can be completed. What AI fluency can do is okay, well, what if we were to iterate on all the possible ways that we can complete this process? And we're no longer constrained to maybe just your five best scenarios. Maybe you can have this technology iterate an infinite number of different outcomes for you. Call it the Doctor Strange multiverse of possible outcomes. So if you now know that you have a tool that can give you uh a glance into this multiverse of infinite possibilities, that is then how you can start getting some insights on, well, maybe there's a different way that we can sequence the work, maybe there's a different way that we can utilize our human capital into doing things that allow us to get on that most optimum path to the outcome that we're looking for. So it is a bit of a vague way for me to describe, but this is maybe describing more of a mindset, not necessarily well, here are three courses that you can take, two podcasts that you can listen to, and one university course. Those will still be there because there still is the uh pragmatic, tactical approach to learning the tools and capabilities. But what's going to be harder to learn, particularly for leaders, is that sensibility that if you had the opportunity to look and see an infinite number of outcomes for your particular business, how it is that you can get onto that particular path. Trevor Burrus, Jr.

SPEAKER_00: 30:13

This sounds like how the Avengers movie happened. But it seems almost like a cheat code. If you can start to unlock all of the potential scenarios, probabilities, and really start to model out some of these things without incredible amount of cost. You know, the the some of the big consulting companies have built their entire empires, you know, they're around parachuting in teams of people to analyze these things. Does that mean that you think people are thinking too small when it comes to AI and when it comes to some of the iterations that are possible and how that would change their businesses?

SPEAKER_02: 30:50

I actually think they're thinking too big. Okay. I think a lot of times we get pulled into a lot of the hype of what these two tools can do. We get enamored by these large valuations, these watering, eye-watering valuations of what companies are able to be worth and what they're able to raise all in this space. All this talk about data centers and but really what organizations need to start thinking about is what we had talked about previously. Where's your data? So it comes back to some fundamentals. And so if you spend too much time thinking about, well, we just need to use these tools and that will solve all of our problems, that's a little bit short-sighted because then by the time you practically go and try an AI prototype or pilot project, then very quickly you'll realize that your data is the issue.

SPEAKER_00: 31:44

Interesting. So it still goes back to those foundations that is what is your data integrity, what is your data fluency, and what is your data governance process to be able to then contemplate what are these really interesting things down the line.

SPEAKER_02: 31:58

Absolutely. And so the relevance of a community of practice of professionals seeking to upskill themselves in data analytics and data science is more important than ever. And so I know some people worried that, oh, they haven't jumped onto the AI bandwagon soon enough. Well, it is just a bandwagon. Like a lot of other bandwagons, a lot of it is propelled and fueled by hype. But if we peel back a lot of that and we just see what is the foundational pieces, well, it is already taking what you know about data in your existing roles and just being to amplify that a bit more, learning some of the techniques so that you can have a better line of sight to all of your data, uh, learning some practices that allow you to keep your data clean going forward so it doesn't all of a sudden mutate into something that you can't use anymore. And then finally just having some leadership by being able to pull insights from your data and translate that into insightful, compelling uh call to actions for the decision makers in your organization.

SPEAKER_00: 33:08

I love it. I'm gonna have to write all those down here. You've obviously been working alongside a lot of leaders that have gone through this process, whether it be through your community or through your career. Are there often surprises that they come across with or they come across when they're at the tail end of these projects and they they come out the other side either achieving something they didn't realize or something that was bigger than what they thought they were going to do? Like are there surprises that they come across?

SPEAKER_02: 33:36

I think there's always surprises when you go on a particular journey or adventure because you can try to plan as much as you want, but there's always these unknown variables that end up showing up. Uh certainly I think that there's uh no lack of shortages of um bad surprises, and often that is really due to the fact that, well, one, your data just wasn't in the right state that it needed to be in. But also, we're still constrained to a very conventional way of running projects and how these projects are sponsored. So you may have a champion for a particular initiative that is really passionate about this and was able to secure the funding for it. Okay, well great. The project progresses, but as it is in a lot of organizations, leaders move around, people move around. And so perhaps the new leader who is now in that position doesn't have the same vision or the same passion. And so by the time the project ends, well, guess what? There's no more funding for it, and then it just dies on the vine. The other consideration is that sometimes there are great surprises, and so a thoughtful leader would anticipate that if one of these great surprises comes up, what can I do to ensure that there is enough support and inertia to progress the project? And so this then becomes sort of a CapEx OpEx conversation. Um, so a lot of funding is secured just for the project itself, but there is no consideration of, well, how do we then scale this so that it's more enterprise-wide and becomes more of an operational expense rather than a capital expense?

SPEAKER_00: 35:22

Interesting. And so you you brought up as well the organizational design side of things. Um you shared some really interesting thoughts the other day around how data, how AI, how some of these emerging technologies are really gonna start to reshape organizations themselves. Are you comfortable sharing kind of some of your thoughts around what changes organizations are gonna have to make in order to really ingest some of these technology changes?

SPEAKER_02: 35:52

You and I were the benefactors of many decades of business studies of what an optimum organization structure looks like. Uh and all of it would have started in sort of this command and control type of structure where you have a lot of employees or a lot of workers trying to achieve certain outcomes, but you want to make sure that you have the right layers of control so that one, leadership knows what's happening across a broad swath of uh locations of where they're operating. Um, but that there's also a standardized way of communications and of executing on the particular work that's done. And you and I we can agree that probably for the last five, six, eight, ten decades that this approach has worked. So, what happens if all of a sudden the structure itself consists of workers that are human, so flesh-bearing humans, but also digital employees, so GPU-bearing workers as well. Because we are now getting to the point where AI is falling under this term agentic, meaning that you have these particular AI tools that are really good at doing specific things. And so if you can somehow group them together so that each AI tool is responsible for one particular part of the process as an agent, then you can actually have a whole team of agents working on a particular outcome. And so we're probably not that far from a future where a manager or a supervisor will have to manage not only these flesh-bearing workers, but also these GPU-bearing workers as well. And that would be a whole new leadership skill set. If you were to enroll into a course in HR, they don't teach you. Well, how do you manage a team of agents? And there could be two agents, there could be 2,000 agents. And so some thought needs to be put into is there also a disruption not only in the tools that are affecting the workspace, but the way that the workspace is being managed and performanced?

SPEAKER_00: 38:07

It's really interesting. And you know, by the time this releases, it'll actually probably be out in the world. But I have a an article that's penned right now that talks about management as the new leadership. Because the idea of motivation, the idea of incentives are different and very different when you think about a human flesh-brained person versus a GPU that um you know just wants essentially compute and energy as its inputs. And and it's really going to change those um you know meta-skills and functional skills of of what does it mean to be a manager.

SPEAKER_02: 38:43

Absolutely. And uh it's exciting, um, maybe a little bit uh concerning as well, too, uh, because I think also we've seen humans are good at really messing things up, right? We kind of again become the victim of our own success, or we let our own broken nature get in the way of things.

SPEAKER_00: 39:03

Oh, definitely. And so speaking of our broken nature getting in the way of things, are there some pitfalls that we should be watched out for? And I recognize we could probably have an entire episode based on the you know the scary things and the T9000 uh future in front of us potentially. But if we think short term and really tactical, what are some of the things that we should be really deliberately talking about or thinking about when it comes to how we're responsibly integrating AI into our businesses?

SPEAKER_02: 39:31

We're in the time of human history where we have access to so much information and knowledge. Uh another very conventional structure that we're used to is our academic institution. Um, and particularly how do I signal to you, Keith, that I've got certain skills? Well, I have perhaps a piece of paper that says that I've graduated from a certain school. But nowadays, people are able to upskill and learn new things really at the instant of a mouse click. So we have all these massive open online courses, we have micro learning that is out there. But yet there isn't a framework or a schema that both an employee and an employer can use to say, well, this is actually what I'm capable of doing, because I was able to upskill myself in these certain ways. All I've got to show for you is just that piece of paper that some institution somewhere has decided to write my name on it to say that, okay, I've got a bachelor's degree, or I've got a master's degree, or I've got a doctorate. So I think the pitfall is if we still allow traditional frameworks to try to signal where there is capability, we may end up missing on human capital and talent that's right under our noses.

SPEAKER_00: 40:53

100%. And I think you bring up a very interesting point too in terms of the education institutes or the educational institutions, because a lot of our educational system is built around knowledge synthesis, remembering and being able to re-articulate um facts, figures, dates, um, theories, calculations. How does your role, and you probably have a very interesting perspective in this, as an educator start to change when the ability to recall and synthesize information is no longer the prerequisite, but there's a huge responsibility and opportunity still for what we do with that information?

SPEAKER_02: 41:34

Trevor Burrus And this goes back to I think the distinction between literacy and fluency. It is more efficient to evaluate someone on their literacy skills. Were you able to read up on these concepts? Are you able to memorize these things, and are you able to successfully answer a multiple choice question? That demonstrates literacy. Fluency is the ability to take all that knowledge, synthesize it, and communicate in such a way that it is compelling. Compelling in such a way that maybe it then supports the changing of policy, that it can instigate new regulations, uh, or even just cause someone to pause and maybe change their perspectives a little bit. So to me, the mastery of literacy is what leads to fluency.

SPEAKER_00: 42:26

I l I love that topic, and I might steal it, or the the idea of that literacy versus fluency, and it's really starting to become clear to me around kind of where the big difference lies there. So as a as someone that is um you know actively an assistant professor, does that change the way that you're teaching?

SPEAKER_02: 42:43

What it does is reminds me that the students are currently in this really odd position. Yeah. They are still having to adhere to some of the traditional code of conduct that the institution has around plagiarism, around how it is that they do their work. At the same time, these students are about to graduate into a work world where employers are expecting them to know how to use these tools. So I actually feel really bad for the students for having been pinned into such a tough situation. But I also see that my role there then, particularly as more of a practitioner-focused instructor and educator, is to help them to start building some of these AI leadership skills. I know very well that they can go and learn how to prompt appropriately in some of these tools. But what they're not being exposed to is, well, how do they deal with change management, you know, changing org structures and being able to implement AI in a responsible and ethical manner. So I actually think that they have the opportunity to gain a competitive advantage versus those that are already in the workforce.

SPEAKER_00: 43:58

Interesting. And so this might be a little bit outside of your expertise, but I'd be curious about your perspective on it. You know, I was at a future of work summit um last week, and there was a lot of talk around youth unemployment, around kind of the changing demographics of the workforce. What's the narrative that you hear your students talking about, as well as some of the employers that you work with, in what's going to be kind of happening to that Gen Z demographic as we're building up a um future leaders, but also as they're getting their super valuable experience on the front end of their careers?

SPEAKER_02: 44:35

It is tough. Like there are uh economic reports that will suggest that those between the ages of 15 and 24 are in a very, very challenged uh employment situation right now. And we are already seeing that a lot of the initial use cases for a lot of these generative AI tools are tasks that would typically be taken on by new graduates and interns. So there's a couple of headwinds facing these students. The jobs that would normally be there for them to graduate into are being taken away by some of these tools. In addition to that, they're coming up against a sensibility that doesn't really allow younger workers to be able to explore some of these tools. A lot of the tools are still pretty much all constrained to the point that it's not really that useful. So then the question is: well, what is the place for these graduates? They have a lot of new knowledge, very recent knowledge, but yet they're not provided with the opportunity to be able to showcase that. So I'm really delighted to be in a city where the focus on technology and innovation is so high, with organizations like Platform Calgary providing essentially a lot of support structures for those that are wanting to consider entrepreneurship or starting up their own companies. This is really where I believe that students and new graduates will thrive. Not trying to find their way into a traditional organization that's been around for many decades and are already codified in their specific ways, but to be part of an initiative or organization where everything is very agile and very, very fresh.

SPEAKER_00: 46:27

I think that's a really interesting place for us to start to wrap up here. Before we leave, I do want to give you a tiny bit of airtime to just share Untapped Energy, what you're all about, and maybe some of the things that you're working on, um, so that our listeners can connect with you if they're interested. So if you wanted to give the 30-second elevator pitch on untapped energy and some of the priorities you have ahead, um, I would love for you to be able to be able to share that.

SPEAKER_02: 46:51

Thank you for that opportunity, Keith. Untapped Energy is a federally registered nonprofit, upskilling professionals in data analytics, and data science. We do this through our monthly meetups, we put together tech social events, and then we have our annual data funds that we run. Our focus right now is actually on microlearning. Realizing that professionals typically don't have a lot of time to enroll into even a multi-week course. They don't have a lot of opportunity to access some of the latest and greatest of knowledge. And so a microlearning course is essentially a 60 to 90 minute, very targeted offering, learning experience for professionals on various topics in the data and artificial intelligence space. So for 2025 and 2026, our focus will be on a number of these microlearning courses. Everyone is welcome to join because we're a nonprofit, they're very accessible. Most courses are between$15 and$20. And it only will take about an hour or two of your time.

SPEAKER_00: 48:00

Fantastic. Well, thank you so much for all the work you put into Untapped Energy, Tim. I know that it's a great service for our city, for a lot of the people that take part in it. And um, you know, it is I really appreciate all the work you do there. It's amazing to see and very, very needed. Um thank you as well for coming onto the podcast today. It's been a phenomenal discussion. I feel like we could probably continue to talk um for for many, many more hours. But if someone wanted to connect with you and learn more, if they wanted to pick up um the conversation or maybe they wanted to pick your brain onto something we haven't fully unpacked today, what's the best way for them to connect with you?

SPEAKER_02: 48:36

Yeah, absolutely. I think LinkedIn is the best way to connect with me. I can provide you with my LinkedIn. I'll put it in the show notes. Yeah. Untappedenergy.ca is the website for uh the organization. If uh folks want a sense of the things that we've done in the past and some of the things that we're working on. And absolutely, you're right. Uh I think we've just scratched the surface on something that continually moves at breakneck speeds. Um but this is where I believe the power of conversation is so important. It allows us to be able to pick each other's brains, share perspectives, and kind of have those mini aha moments, like, oh okay, never really thought of it that way. Uh and this is how I think we will continue to embrace something that is very uncertain to us in a way that does result in something that is good for us overall.

SPEAKER_00: 49:27

Phenomenal. Thank you so much for coming on today, Tim, and can't wait to keep talking.

SPEAKER_02: 49:32

Okay, sounds great.

SPEAKER_00: 49:34

If you've made it this far, like and subscribe on YouTube or follow and leave a review on your favorite podcasting platform so you don't miss any future episodes.