Episode 117: The One Thing About AI You Can't Look Up: Sheryl Barbosa on Fear, Facts, and Your Own Team

Erik Martinez: [00:00:00] A while back, I did all the visible things for AI adoption inside my own agency. I subscribed to the tools. I attended trainings with my team. I put somebody in charge of coordinating our AI effort, so it wasn't all on my plate. And it stalled, not dramatically, just quietly went nowhere. People did their own thing or they did nothing at all.

It took me a while to figure out why, and the answer was me. I wasn't showing anybody my own work or showing them how it could help their work. I wasn't demonstrating the value. But to be honest, I was also learning at the same time. And what I was assuming without ever saying it out loud, was that subscribing to the tools and running the training would be enough to get people learning with the same enthusiasm I had.

I have been thinking about how many other things in my business are like that, beliefs that I hold, but I haven't revalidated to see if they're true. The questions I get about AI now aren't really about whether it works or not. They're about how we should be using it. Do we modify jobs or replace them? [00:01:00] How do we protect privacy and still get the analysis we need out of our data? And lately, the one I keep getting is about the environment. What's AI's real impact, and is it worth it?

The twenty twenty-six State of AI for Business report shows that leaders think AI is essential to the business, and yet their people are still uneasy about it. Then the twenty twenty-six Edelman Trust Barometer says that trust is getting more local and more fragile, and that people are looking to their employers and their coworkers for signals about what to trust. Put those two things together and it's a strange place to be. So the question I keep coming back to is, what does a leader do when concerns like that land on their desk?

I called Sheryl Barbosa, who does this for a living. She's an environmental scientist and a sustainability expert, and she sits on her own city council, which means when a data center goes up outside of town, she's in the room with the neighbors.

She was on the show about a year ago, and enough has changed since then that we needed to do it again. Sheryl, welcome back to the show!

Sheryl Barbosa: Thank you for having me again. I had so much fun the first time. I'm so glad to come back.

Erik Martinez: and that is probably the longest-winded, intro you've ever heard. before we get started, Sheryl, would you share a little bit about you, what you do, and, your expertise?

Sheryl Barbosa: So I have been at my career for over 20 years now, so I have a nice variety of background experience. The core of the work that I've done throughout my entire career is being an environmental scientist. So, consulting in strategic planning and risk assessment, risk management when it comes to the environment.

So looking not only at how do you plan for the future, what works for your organization, but also, what are the risks, and evaluating how can you best avoid those or reduce, what the outcome is of that risk if there's no way of avoiding it. and throughout my lovely tenure, I've also done [00:03:00] things like I founded an organization and business called Sustain Cities Network, which works with local, connects local businesses and customers on a very small scale and city by city. I also am currently on city council for my local small city here in Travelers Rest, South Carolina. and I've, I serve on the advisory board for, two different nonprofits, which I get to fill the role of being a mouthpiece for the planet and advocating, so looking at sustainability and having those discussions when it comes to, technology and the space industry as well.

Erik Martinez: Sheryl is also an avid traveler.her work gives her the advantage to travel around the world and see some of these problems firsthand, and challenges and opportunities, in places all over the world. So Sheryl, as an environmental scientist, you're trained to look at systems, trade-offs, unintended consequences and prevention.

When you hear people raise concerns about AI's environmental impact, what do you think [00:04:00] business leaders are missing because they're looking at the technology too narrowly?

Sheryl Barbosa: So I think, you know, when business leaders are looking at AI's footprint, they're looking at it as like a single transaction. so, you know, as a scientist, when we look at it, we kind of look at it in a cumulative way, and not only what has happened in the past, but what is slated for the future.

and so essentially what's happening is there's this mistake that, of comparing like one Google search, let's say, the equivalent of a Google search, instead of, you know, asking like, "Okay, what happens when all of us do this together?" and, you know, that's something that, there's legitimate concern around.

but also there's sometimes this misunderstood fear. Certain adages will stick, whether, it was a truth once upon a time, but things have now changed, or it's just something that maybe the media got ahold of and they pulled out this weird headline and it took off and went viral. and, you know, I think that that plays into a lot of businesses and [00:05:00] organizations, whether it's a leader and their direct concern, or it's a concern that's coming up from the employees and the individuals that they manage and work with, and having to address that or have a discussion around it.

So I think that it's legitimate concerns and questions, but I also think it sometimes oversimplifies things as well.

Erik Martinez: What do you think the difference is between a legitimate concern and a misunderstood fear? Because I think, sometimes We assume that that's a concern or a complaint, and it's really just a fear about the unknown. And so how do you help leaders discern on that point?

Sheryl Barbosa: So typically, especially when it comes to AI, a lot of the concerns, environmental concerns can be, whether it's water usage, energy usage, and just in general impacts on the environment. And from just a human perspective, how does it impact local economy, the neighborhoods [00:06:00] surrounding this particular location and data center, which are all, legitimate concerns.

do have an environmental piece that plays into them, whether it's sound or things like that. And a lot of those, because they're data-driven, you can debunk them. you can do your research into them and really truly dig down into, what that means. And if you don't have the time or don't wanna do the research, you can look towards experts like myself, or other individuals in the space that are already doing this research and are already paying attention to what's out there, what the cutting edges are, and what's coming down the pipeline so that you're able to take a look at, well, if the concern is, we're concerned about power usage, that these data centers are going to take up all this power and either drive prices up or have a cause to produce more power that's not necessarily renewable or safe for the environment.

And that was true, and it is a fear, but as technology has been changing, [00:07:00] and we can touch on this, more specifically later on, it's, not the case moving forward. And it's starting essentially pretty recently, within the last year or so with these technology, advancements in the physical components of AI and data centers and whatnot.

We're starting to flip that switch in another direction. And so being able to have that data to speak to your employees that have these concerns is really important, and that is where it takes a legitimate concern and addresses it, or helps debunk that kind of misunderstood fear that maybe isn't true any longer.

Erik Martinez: I'm curious. outside of going to an expert like you, are there, some easy-to-access resources that have some of this data that a business leader could point to to have that conversation when somebody says, "Hey, I'm really uncomfortable with the cost of AI in the environment as I know it today?"

Sheryl Barbosa: Yes. So right now there isn't necessarily one source or [00:08:00] website to pull from because there's so much data coming in and there's so many different... So I get a lot of my information from, credible resources that are, have been proven, especially over time. So studies done by the United Nations or different universities.

Well, especially studies done more so in Europe because they are, they're a lot more robust, and a lot more, for me, environmentally focused. right now, currently as things have, are going on in 2026, a lot of our environmental regulatory organizations have, lost their footing. or, you know, their funding has been reduced, so they're not quite as, what- what's the word I'm looking for?

It's a little bit of a struggle for them to not only do these studies, but be able to talk about what they're finding. So I tend to go more towards sources that are based out of global and getting data from other areas, including Europe , and also through studies through various institutions and [00:09:00] really it's Google search.

Taking a look online, and following people as you find them as resources and realize their trust and credibility. And there's a lot of people on Substack talking about this as well

Erik Martinez: those are great options to, to look at. So let's pivot to the conversation on trust, because trust, I think, is pivotal in this whole discussion. so the Edelman Trust Barometer talks about declining shared trust and the need for trust brokering, helping people with different concerns find common ground.

Do you think some of the backlash against AI is really a trust signal? In other words, are people asking, "Can I trust the people making these decisions?"

Sheryl Barbosa: Yes, and that is more... It's interesting because it's, it's more hum- it's more focused on human to human interaction versus the actual AI itself, right? I think we all know that AI is still very imperfect, and I hope that we actually keep that mentality. AI has been [00:10:00] around for, a very long time.

I think a lot of people don't realize that. It was just in the recent last, you know, few years where it became very predominantly used for the public and democratized in that way where we can all use it. But I think that's the biggest concern, is just it's not so much when people are like, "Well, I don't trust AI." It's not so much that you don't trust AI, you just don't trust maybe the people either using it or creating it, and what their morals and values are or what their intentions are behind it. So that is something that I feel like, especially within an organization, regardless of who is speaking, whether it's a manager with employees or it is a executive team and a level of peers all speaking with each other, the most important piece is remaining, encouraging and continuing ongoing conversations. Creating a space where people feel safe to be vulnerable and speak up about their concerns, to feel heard, and just having this [00:11:00] ongoing discussion of, okay, change is happening no matter what. So you can choose to embrace that change is happening and help shape what that looks like, or the other option is to try and, ignore it or fight it.

And that change is still gonna happen, and the drawback with ignoring it or fighting it is that, you're gonna be less prepared for when it shows up, and it's gonna kind of be forced on you. You don't really have a choice at that point, and then it feels even more, anxiety inducing because you haven't had the time to prepare or wrap your mind around it.

So it's good to at least start educating yourself and learning about it, even very passively, and understanding, where it's come from and it's not just this brand new thing that suddenly cropped up out of nowhere.

It's been around for a very long time.

Erik Martinez: Yeah. So what I'm hearing you say is probably the best way to establish that trust is having that open dialogue. Like, " Here's why we're using it. Here's what we think the benefits are gonna be. We understand it's not [00:12:00] perfect."

And I agree 100% with you. I play with this stuff all day, every day, and it is sometimes quite maddening how not good it is from time to time, and yet it- there's amazing things that can be done. So, you know, I think, one of the things I think some business owners and leaders are struggling with this concept of like, I don't know if that they necessarily feel they need to apologize, but you're almost kind of sometimes getting pressured to, say, "Hey, I'm sorry, but we really have to use AI."

I think, we need to strike that balance between, " We're moving forward with the business. We want to utilize this technology, to help us become more efficient and, more profitable." If we're more profitable, it's better for everybody on the team, right? but when you're talking to your customers, let's say your organization's all on board and everybody's happy using these tools, and [00:13:00] yet you have customers who now are kind of sitting on the opposite side of that fence.

How do you open that conversation You have a customer who comes to you and said, "You know, we're really, really concerned about the use of AI, not necessarily for the business purpose, but we do have this very strong concern about its impact on society and our environment."

How do you frame that conversation?

Sheryl Barbosa: Well, I think the number one thing to remember in those situations is don't be dismissive The best way to approach something like that, because it's new, right? It's new to all of us, and it's changing so rapidly that to have the answers, you know, concrete answers, is almost impossible, right?

But to be able to speak with clients and say, "Here's what we know. This is what we're watching. We appreciate, you know, your concerns, and please keep bringing them to us because we do wanna hear from you and what your concerns are." It helps us. Like, we don't know what we don't know. and also, you know, from a perspective of [00:14:00] not only environmental impacts, but also, you know, privacy and concerns around how data's being used.

It's one of those things that, you know, assuring them that you're taking the steps you can to protect them, to protect their privacy, which I think we all are, because cybersecurity, and even before AI and data centers, that was always a concern because our digital footprint is everywhere now.

As somebody who travels internationally, they just started implementing these systems where, to go through customs, it is an efficiency thing, but it's also, you know, it is more foolproof from a security standpoint, where, you know, when I walk in, it is a double gate with two doors.

So the first gate, I have to put my passport on, and it scans my passport, which is, that's fairly normal, I think, you know, even when you're going domestic flights, they're scanning your license, when you go in. But the second one is a set of fingerprints. You have to put your hand on this digital keypad and, you know, they're essentially saying, like, you are who you say you are.

It, it, that's really what it is from a security standpoint. And [00:15:00] so, right, all of that stuff is happening every day to all of us. But you also as a business leader don't wanna say it in that way to your clients. And you want to foster through it all, whether it's with customers and clients or employees or among peers is, again, having that conversation of like, "I hear you. I see you. I'm on the same page. I have the same concerns, and we actively are, you know, monitoring this, staying on top of it, and being able to make conscious choices and decisions."

And, it's very early development as far as efficiency in these data centers, right, physically, the amount of power and water and whatnot they're using. And so there is definitely more of a push, especially if the company is conscious on their environmental footprint, and that is something that they promote actively, and they don't wanna be seen as greenwashing, being able to offset that by, directing any kind of donations or givebacks or things like that, like giving back to organizations like 1% For the Planet- Or, explaining, even just talking about like, yes, we use AI, but we're using it to help [00:16:00] refine our logistics.

So we're reducing our carbon footprint through our shipping because we're making that even more efficient. So you're offsetting that until these data centers really truly get to where they're going to be, which is something, you know, we can touch on a little bit later because it's, it's actually what's happening currently and what the future holds in the very near future is actually really promising, and I'm, I'm actually very excited about it.

Erik Martinez: Yeah, it's incredible to watch how quickly this technology is changing. And because it's changing so quickly, what I'm hearing you say, kind of the through line is the premise that we have started with, even from a year ago, that fundamental premise has changed

Sheryl Barbosa: Yes

Erik Martinez: in less than a year because there's new technologies coming out and new ways of doing, running businesses and making them more efficient and more green at the same time, which is the perfect segue into our next topic because, [00:17:00] you know, last time you were on the show, we talked a little bit about what businesses can do to help manage their resources.

You know, we talked about routing in your logistics program. We talked about managing your inventory and redoing your packaging and waste and operational efficiency. Where is your current thinking on what businesses can do? has it increased, or is it still focused on, hey, there's some fundamental things?

I think I know the answer, but I'm gonna let you tell me, and then I can tell you if I was wrong.

Sheryl Barbosa: Oh, absolutely. So, you know, everything I said in the previous podcast, about the physical logistics and the, you know, especially for people who are producing products still stands for sure. But what I like to add on to that too is, so there's a lot of service-based businesses, right? That are using AI and using this technology to help themselves.

For them, their waste is really their time and attention. And so it's not necessarily a physical resource, but even so, that time and attention can [00:18:00] be best used for, you know, what you're making up for inefficiencies, right? And optimizing things, optimizing what you're doing every day.

Here, I'll give you an example. So we have a lot of AI note takers, right? That come into meetings and take notes and whatnot. So before AI note takers, we would, sit through a meeting and say an employee would, be taking notes and summarizing things, and somebody after the meeting sends out a recap email.

That can take, depending on the meeting, anywhere from 30 minutes to half a day. It depends on what was discussed, what's needed, if they have to do more research, whatnot. So not only are they, from an efficiency perspective, their time and attention is going to that versus something else that would be, more advantageous for them and the business.

But also, there's resources wasted, physical resources of them with power, on their computers, and so there's a physical component to that as well. So now we have these AI note takers, and they're able to spit out something within minutes. And yes, we go [00:19:00] back and, read through them and maybe add to them or whatnot, but that's reducing all of that down to, that recap.

some people just go, "Oh, that looks good," and sends it out, and it takes a minute or two, and maybe they take 20 minutes to take a look at it and correct some things, make some edits, maybe add to it, and then it's out the door. So you're saving that from a business perspective, and that then turns into, stewardship happening because of AI, an environmental stewardship because you're trading one for the other.

So the individual within their job is now spending their time on things that are more effective, things they actually enjoy doing. So they're gonna be a lot more, efficient and also just excited about their work. And it, overall increases the morale and the health of the business itself as well.

Because AI isn't necessarily here to take our jobs, right? it's taking certain tasks away from people, but what AI is good at is very repetitive, basic type tasks. The creative stuff, I think we're all well aware [00:20:00] that it is not very good at that. Even if it is good at, producing a video, you still have to babysit it and go in and tweak it, and it's not something you can just pop off.

And maybe eventually in the future it is, but not really, because it's not a human brain. And it's just not wired for that. But it's very good at being structured. when you're keeping it structured and keeping it on rails and giving it very specific instructions, it can do that same task over and over and over again, and very efficiently.

So it's usually taking away those tasks that most of us don't honestly wanna be spending our time doing anyways.

Erik Martinez: There are a lot of them.

Sheryl Barbosa: Exactly.

Erik Martinez: it's funny that you say that because I was telling somebody the other day that I very rarely go into my spreadsheets anymore. I look at my spreadsheets, but I don't do the work of creating them anymore, the AI actually creates all the data, validates all the formulas. I verify all the outputs, but I never actually go in and really do much of that work anymore. I don't put in the [00:21:00] calculations. I don't do any of that because I use the AI. And again, I'll do the validation, which is eminently more fun than actually writing all the formulas, and creating all the macros, and doing all those things.

You know, it's also interesting just in terms of what businesses can do. I'm gonna take this a little bit to the personal side, like we were talking pre-show. I'm in a temporary office scenario because I'm retiling this room. I did all the planning with AI 100%. The parts list, the how much tile I needed to buy, how much, mortar I needed to apply to the floor, all that.

Now, yeah, it's not perfect, but it gave me... I went into the hardware store. In the past, I would've had to talk to the guy for an hour, an hour and a half. I went in, I had 90% of my answers, and even though some of the materials weren't exactly the same as what the AI gave me, dude's like, "Oh yeah, here's what you need.

Here's what you need. Here's what you need," and I was out in [00:22:00] half an hour. That would never have happened before, right? And I think that's what you're kind of saying is, like, there's some trade-offs in terms of the, of the efficiency of doing that. Now that employee had another...

You know, instead of spending an hour and a half with me, now he can go help another customer.

Sheryl Barbosa: Yeah. and from the sales perspective, he's able to make more sales because he's able to obviously service more people, but also it's a lot more of a, oh, you've now been educated. He might say, "Oh, I have this other product that does the same thing but is maybe a little bit better, cheaper," whatever the case may be.

And so you're already going in there knowing, "Okay, this is kind of what I need," and the sale's kind of half made. But versus before where you're like, "I don't know. I'm gonna go shop around," and you might leave. and so yeah, it definitely helps in a variety of ways that we don't necessarily think about right off the top of our head

Erik Martinez: I can give you another example. our refrigerator died and we have a really tight space that it fits in. [00:23:00] There's only a handful of refrigerators this particular size, and if I had not done the measurements and used AI, I would not have known which models to go after. I would've ordered one, and it would've been too small or too big, which is another form of inefficiency.

And I think you can start taking that approach to your work, right? We were talking about solar panels, and you taught me something pre-show that I didn't know, like, that you could go get used solar panels at a reduced or a fraction of the cost, have them installed, and they're still, like, 80% efficient, and that's a way a small business could potentially save some money.

I think some of those things are just not obvious to us because we don't live that every day, so I think those are really good.

Let's go ahead and talk about this data center thing, because I think it's a hot topic in our society. I think there are some legitimate concerns, about electrical prices.

With the price of everything going up, you know, our utilities, [00:24:00] our groceries, everything's going up. Let's have this conversation about why are data centers viewed so negatively at this point in time from your perspective as a scientist in this field?

Sheryl Barbosa: So there is a lot happening right now. As you can imagine, like AI, it is a rapidly changing situation, and decisions are being made, every day, probably as we speak. And so same as AI and a lot of other things that we end up, as a general public, experiencing well after it's already been developed.

There actually have been data centers around for a very long time. and it is something that there's, obviously because they've been around for 20 years, the internet... I like to say I'm 43, and so I kind of, I grew up as a kid without the internet, and Windows 95 was kind of the beginning of that, right?

Because prior to that, you didn't really have a need to... You had a computer at home maybe, but, like, there was no [00:25:00] internet. Dial-up started. So, like, 1995 was really kind of what I like to call the start of that too, generally. But, you know, so some of these data centers, there was always data that needed to be held in servers somewhere.

So some of these data centers are quite old. And so the concerns that are legitimate concerns, 'cause they do exist, but, from previous old data centers, like, yes, they're not that efficient, and they can be loud, and there's a lot of need in order to cool them and this power usage. But what's happening now, and this is where it's, like, a legitimate fear is kind of starting to shift more into a misunderstood concern, is that they're becoming insanely efficient .

I mean, truly think about it. Even just through the lens of businesses. So by nature, they're gonna want to maximize their profits and reduce their overhead costs. So they're gonna continuously be looking to make their physical product, these data centers, more efficient, but they're gonna wanna figure out how to make them mobile. And then they're gonna go to the point even, which is ha- So [00:26:00] right now, what's happening, It's not super known to the public. There's nothing secretive about it. But, what's happening is there, because there's been such pushback, rightly so, for people on data centers and building data centers in these areas, what's happening is they're shifting now.

Not only are they maybe physically putting them away from where people live, and part of this is efficiency. You know, dropping a data center right in the middle of a city, you have a lot more pushback than if it's in the middle of, you know, a couple hundred acres with literally nothing around except for maybe cows.

Who knows? So that, there's something like that that's happening. But also, in the interim, while they've been kind of navigating that political landscape, they've figured out that they... And they've become so efficient that they're able to put them in, shipping containers now and make them modular So they can stack them.

And so they have them in shipping containers on ships in the ocean. They're tied to different Skynet, Starlink-type things where it's satellite-based versus wired [00:27:00] in. So they're on vessels out in the sea or, you know, they're able to put them in smaller clusters in places, which also helps reduce stress. Almost all of our electrical infrastructure, the power lines we see, whatnot, the big ones that are transmission, doing transmission are old.

They're old. They need to be updated. They've needed to be updated for a long time. So this massive draw on these systems that would be needed for a much larger center, are what, you know, that was the concern, right? So These are tech companies. They're like, "How can we, essentially make this the most smoothest, most efficient situation for us so we don't have to spend two and a half years doing a electrical study before we can even build?"

So they're making what they call behind the meter. They're figuring out how to do this behind the meter, so they're making smaller modular centers. They're also making them more efficient with their learning, and this goes for anybody who's technical, even cars. You have antifreeze in your car. It's not water.

It's chemicals mixed with water, right? Because those chemicals do a much [00:28:00] better job of cooling than water does. And so they've realized that. So these, a lot of these new centers and these new technologies don't use water. It's a closed loop system, and it is a combination of the two, or it's just a entirely different, you know, I don't want to say chemical for people to freak out, but it's, it, it's something that can be used to help with cooling these centers.

So that concern of, "Oh, my God, they're gonna be pulling all this water from our drinking water," is becoming obsolete. Same thing as they figure out, what the future holds is they, yes, they produce heat. Okay. Well, how can you use that heat to create energy? Because if you can then not only create energy from these data centers, either you can help power yourself, or you can sell that back to the grid.

If you have a giant warehouse with an empty roof, why not put solar panels on it? Helps shade and cool the building. You're helping power your center, and you're also selling, so you're making money that way as a data center. So what the future [00:29:00] holds is, I think it's gonna be a wide variety of solutions to fit things as needed, but I think that there's gonna be a level of efficiency where this power draw and these power concerns are gonna go away.

And the same thing with water, and that environmental impact is very much going to be, you know, be eliminated or very greatly reduced. And also, I think we're gonna see a lot more of these popping up on contaminated sites that wouldn't have been able to be used for anything else. Because most of the time when that happens, the solution is we'll treat it, but we can't have anybody living there and dealing with businesses there. It's just pave it over and put a warehouse. That's always been historically the thing is pave it over, put a logistics warehouse there. So I think you're gonna see some of that happening. But with data centers, and the beauty in that is then they can sell back and help the grid, help your local energy supplier or even potentially, Give back to the community in that way.

They can set that [00:30:00] energy production to feed directly into the surrounding community, and then that can be earmarked for low-income individuals that are struggling to pay their bills to help supplement that. So there's a lot that can happen through that perspective with AI and these data centers, and that's something too, I think, that moving forward, even businesses, similar to being able to choose, like, "Oh, I'd rather know that I'm using energy from renewable sources than coal plants," I think similarly that's gonna kind of become a thing.

Like, this will be commodified, and it's like, "Oh, we're a business, but all of our data centers only use renewables."

In an ideal world, that would be great, and really, it's a really interesting concept in where things are going because those organizations, as businesses, want to maximize, and they will. They absolutely will.

It'll definitely go that way because of the pushback they're getting on all fronts, not just public, and i- concerns which are very [00:31:00] important, but also just the physical logistics of trying to... When you're consuming that much power and water, there's a lot of permitting, and it can take years for that permitting to move forward and go through.

And so they can buy a piece of property in your town, but it might be three to five years before they even start construction, and there's a lot that can happen then. So I think they're, you know, we're gonna see a very big shift in how this looks and, you know, it could even go as small as, you know, we see cell towers on the tops of buildings, right?

A lot of hotels or skyscrapers, they lease to cell companies for cell towers. There might very well be similar situations like that where there's these micro data centers in containers on top of these, these spaces or, you know, down on the property around there, utilizing and becoming even more efficient, and then that all just kind of get links together, you know, whether it's through wired connections or satellite connections.

So it's a very interesting landscape.

Erik Martinez: That is very [00:32:00] interesting. Your whole topic here has brought up a segment I listened to on Marketplace, on the Marketplace report by American Public Media, and they were talking about modularization. Modular nuclear plants that are mobile, which would allow you to set a data center anywhere

Sheryl Barbosa: Mm-hmm.

Erik Martinez: and have its own power source.

And, we've been using modular nuclear in this country, at least for military applications, for lots and lots of years. But there's now a commercial reason to make... As you've talked about, our infrastructure is getting old and needs to be replaced and replenished , but modularizing our energy infrastructure is actually probably better in some ways, and some of these new technologies, could help do that. That's kind of cool.

Sheryl Barbosa: Yeah, out west they're doing geothermal. So where geothermal is possible, they're incorporating geothermal 'cause it, two birds, one stone. You're reducing the heat, you know, you're able to cool, and you're [00:33:00] producing power for the, the actual system itself. If it was possible to do that everywhere, it probably would've already been widely adopted.

But it's like other renewables, it's just, it works really well in some places and not so great in others. And so you kinda gotta pick and choose. It's not a one-size-fits-all

Erik Martinez: I really wish I had geothermal out here on the farm, but that's a whole different topic for a different day. Sheryl, thank you so much. I want to be respectful of your time. I'm curious, if you were to sit down and advise, an owner or a business leader for something that they can do in the next 30 days that's really accomplishable, we've talked about a lot of things in terms of, policy, conversations, small moves that they can make to make their own businesses a little more efficient themselves, what would you recommend, that somebody could do in the next 30 days that would be a meaningful step forward?

Sheryl Barbosa: You are not using it at all and there has been zero foundation, is to start that conversation first. and [00:34:00] then that can segue into, general policy, and have it be something that is a little more receptive and open to everybody in the company. To start a discussion really truly is that.

I think a lot of people are using AI, whether it's employees using it personally or companies just kind of starting to dabble in it. So if you're already kind of past that conversation and everybody's aware, but you just haven't really implemented it yet, start with a policy, a general policy around AI.

start kind of getting your ducks in a row, 'cause especially with businesses, there's a concern about privacy, there's a concern about, safety and security with that regard. and I know, a lot of companies, a lot of large companies have their own kind of internal AI. but for the small businesses, it definitely start setting a policy for that kind of stuff, depending on what it is that you're doing.

And, you can even kind of field it out to your employees and send out an email and have them say, like, that question of, what do you think, within your position you could use AI for? 'Cause then [00:35:00] that gives you a general gist of, okay, so from an efficiency perspective, here's a list of honestly, truly the things we might be able to automate and take off people's plates, and it's a good use for AI.

And you'll know. If you don't have a whole lot, maybe you don't need AI. I mean, not everybody's gonna need it. That's not necessarily a you have to have. but it's definitely worth starting those ongoing open conversations about things and start getting policies in place, and have them be malleable.

I know a lot of standard operating procedures are, this is how it is, and there's a reason for that and it's necessary. But you do need to have policies in place, in general as a business owner to make sure your employees are, staying on task and not doing anything with AI that could get you in trouble as a business owner as well.

So starting off with that policy first and making adjustments as you need to go, and then going from there. And I think that, everybody will be appreciative of it, especially as they start seeing how it really helps them out, on a day-to-day [00:36:00] basis.

Erik Martinez: I think that's great advice. Sheryl, thank you so much for coming on the show. Anybody wants to reach out, what's the best way to get ahold of you?

Sheryl Barbosa: we can put my email, in the show notes. that's probably the best way to reach out to me or I'm on LinkedIn. it's my name, so Sheryl Barbosa, but Sheryl with an S. Can find me there and shoot me a message there through LinkedIn.

Erik Martinez: I came into this conversation wanting better information on how to alleviate concerns about the environmental impacts of AI, and Sheryl gave me plenty of it. The piece of information that was most interesting was the discussion about water. A lot of us have been saying that data centers are drinking the local water supply.

And it turns out that a lot of the newer data centers are in closed loop or don't use water for cooling at all, which means the thing that we have been repeating and hearing about on the news had a source that could have been checked. But what I learned from this conversation wasn't about data centers.

It was about my own people. What I had decided at the beginning of this AI journey was that if I subscribed to the AI tools and provided the team some [00:37:00] training, then they'd show up wanting to learn about AI the way I did. And the crazy part about that is that there isn't a source or a study, an expert or anybody on Substack writing about my team.

Meaning, if I want to know whether something was true for my team, there was exactly one place to go, and I never went there. Which is why one of Sheryl's lines has been rattling around my brain since we talked. She tells a worried client to keep bringing their concerns because we don't know what we don't know.

It's easy to hear that as customer service. It isn't when somebody on your team pushes back on what you believe is true. The team has usually thought harder about their concerns on AI than you have. The person who's worried about their job has likely spent more time working out which parts of their job a machine could do than I have.

That's not friction to manage. That's the only research available on the question you actually need to answer on the subject. And when my team didn't adopt AI as [00:38:00] enthusiastically as I did, I was treating that as the problem. But now, I think that's the source of information I need. So if you take one thing out of this episode, ask one question of your team.

What do you think AI could actually do to help you in your role with the company? Don't ask it as a productivity question because you'll get productivity answers. Just ask and see what comes back as data to help you inform the answer to your question. If the answers come back thin, that can be real information too.

Sheryl was direct about that in a way I appreciated. Not every business needs AI in the same way. It's not a have to have. And if the answers come back with rich detail, you can use that data to propel innovation, process, and governance for your entire team.

Thanks for listening to this episode of the Digital Velocity podcast.

Have a fantastic day!