Aired Sep 3, 2026 · 33 min

Listen on: YouTube · Spotify · Apple Podcasts

AI is transforming the utility industry rapidly, but does that mean it’s going to eliminate all utility jobs?

Google’s Global Head of Power and Energy for Cloud, Raiford Smith, joins us to explain why he thinks that’s the wrong way to look at AI. We explore how utilities are using AI to accelerate grid planning and infrastructure construction, what it could mean for engineers and other utility workers, and why even smaller utilities can take advantage of these tools.

Plus, Smith shares real-world examples of AI cutting planning and construction timelines, explains where human expertise remains essential, and offers advice for utilities wondering where to start.

Matt Chester: Every AI conversation in this industry defaults to the same fear. It's coming for jobs. Google's global head of power and energy for cloud says that's the wrong conversation entirely. But how do we know that? Well, in this episode, we ask him. We ask him how AI is changing the workforce, whether it's taking away jobs, how much the AI is hallucinating and making people's jobs harder, and uncovering some big news about Google Cloud's work with a midsize utility and what lessons that has taught them.

When people bring up AI in major infrastructure construction, especially at utilities, often the conversation is focused around potential job loss. But according to Google's Raiford Smith, who's joining us on the podcast today, the conversation should actually focus on something more optimistic. This episode will explore where AI is already changing utility build-outs, how new AI approaches can dramatically accelerate planning and operational analysis, and what that means for an evolving utility workforce, and whether these capabilities can scale.

If you're watching this on YouTube, be sure to like this video and subscribe to the channel for more smart energy conversation. But now let's dive in and thank you for joining us today, Raiford.

Raiford Smith: Thank you for having me. I appreciate the chance to chat with you.

Matt Chester: Of course. And let's just dive right into it. So, Raiford, there's obviously a lot of discussion happening today at the intersection of energy and AI. What would you say those conversations are currently getting wrong?

Raiford Smith: I think three different areas where I think there are challenges or perhaps a new way of looking at things. I wouldn't necessarily say wrong. The first one really is around AIs and AI hallucination. There's a lot of concern about that. I think the second one... is concern over jobs. I think a lot of people are worried about what this means for their roles and for what they'll do in the future. And the third is, you know, the resource requirements and the costs associated with the infrastructure needed to supply and support data centers. I think there are key areas of nuance in each one of those three that actually paint a much happier picture than perhaps what people are thinking about today.

Matt Chester: Okay. That leads me to wonder, you know, you're at Google and so you're kind of at the epicenter of a lot of these conversations. Are there major misconceptions that you see utility leaders and stakeholders having about AI and what it what it does, what it can do, what it means?

Raiford Smith: Let's take each one in turn. So we'll start with the hallucination side. You know, when you think about AI, we get a lot in the press about hallucinations. But in reality, hallucinations or the ability to be creative is a tunable feature within the models. And for certain things like safety or following the law, we absolutely don't want it to be creative. Like it's a strict if this, then that sort of heuristic. And that's where you basically tune that feature down and say, you know, follow the rules. On the other hand, there are times like in our human discussion or in creative problem solving where you want the model to be creative because it's trying to look for things that perhaps haven't been tried before, haven't been done before, or it's trying to understand the nuance of speech. These things, I think, are where AI plays best. And it's really good at those sorts of things. But you don't apply AI, you know, it's like the proverbial, I have a hammer and now everything is a nail, right? You want to use it in the right way at the right time. So that's the kind of the thinking about hallucinations.

On the jobs front, you know, I liken this to, you know, when first we started digging holes and creating things, right? We used a shovel. After a while, we created machinery, heavy machinery to do this for us that could do it faster. And frankly, it could do a lot more than what just one person with a shovel could really do. We didn't stop using shovels. We similarly were able to do a lot more. And so rather than saying, well, go out here and dig the ditch with the shovel, now they're using all this heavy machinery to create all sorts of things.

And I think AI is kind of in that same path, right? We were doing things before, but now AI unlocks a lot of new capabilities for us to either do more work or to do it in a more innovative way. And I think that's the big change here is that people are not going to necessarily be doing the jobs the way they do it today. They may still be doing that job. They just will be doing it differently. And I think that's what AI allows us to do.

And frankly, by the way, if it's speeding up work, it means there'll be a whole lot more work to do. So I think there's a really positive story there for employees is like, there will be plenty of work to do. We won't do it necessarily the same way that we're doing it today. And so training and job placement, those sorts of things are really important. But you have an entirely new way, a new capability that is going to unlock so much for us. And really, that's the way I would look at it.

Matt Chester: I want to dig a little bit more into that unlocking that you just mentioned. When we look at the energy sector specifically, so many of, let's just say, the headlines about AI and energy are, it's either about the amount of power that the data centers themselves use or the ways in which large language models are changing work. And those are obviously big stories, but I'm almost wondering, are those almost more surface level compared with the real depth of what can be done with AI when you really dig into it?

Raiford Smith: Yeah, well, I think the first thing we have to recognize is that the communities that are talking about and having concerns over things like noise, water usage, energy usage, et cetera, costs, like those are legitimate concerns. And I don't want to downplay that at all. On the other hand, you know, it's why Google has a particular philosophy in terms of the way we go about doing much of this, which is first we look for new to the world, carbon-free energy to supply our data centers. And that's really important to us because we recognize that it isn't just buying power. It's also the impact on the communities where we and they live. So we want to make sure that we're doing that in the most sustainable way possible.

The second thing is that AI is itself an enabler for much of the new capabilities to use technology, all of those resources much more efficiently than ever before. We have countless examples inside the company of where AI is doing everything from reducing costs, reducing energy usage, reducing water usage. All of these things are really, really important and it couldn't be done without the use of AI.

And basically, it's a symbiotic relationship here using resources to create better AI and in turn using AI to do these things more efficiently. And so I think that's a story that's been told basically since the beginning of time, from the beginning of economics. Like we're always learning new ways to do things more efficiently. And AI is just the latest tool to really help us do that.

Matt Chester: Yeah, that makes a lot of sense. And so now let's apply that to the broader topic we mentioned in the introduction, which is construction of major infrastructure in the utility sector. Prior to AI, what parts of this workflow are still inefficient and overly manual?

Raiford Smith: Yeah, well, let's kind of break it down into a couple of different functions. The first one is planning, where we're trying to look out into the future and make a decision about what we need to do, what's the right set of resources, what's the lowest cost, you know, to really to go accomplish that. A lot of that is very manual and very slow. Planning models can take hours to days to run. And this sort of length of time to be able to do the analysis means that interconnection queues are backing up. And, you know, utilities may not actually have the least cost solution based on having a limited number of model runs before time runs out and you've got to submit your plan.

And this is where I think AI is actually unlocking some really interesting benefits. First, on the planning side, one of the big constraints is around the interconnection queue. And we look at Tapestry here at Google. Tapestry is able to basically work through years worth of work in a very short amount of time to reduce the paperwork and the validation process necessary to validate new applications for interconnection.

Similarly, DeepMind has created, DeepMind is our R&D division for AI. DeepMind has created a new security constrained optimized power flow model that runs about 30,000 times faster than the traditional model, meaning what would take minutes to hours to potentially days can now be seconds.

What's really important about that is instead of being limited to six or eight or whatever many runs that you would be doing because it was so slow, now you can run and create a probabilistic outcome. You can do a lot more in terms of understanding risk. You can look a lot more simulations and studies than you could ever do before. These sorts of new approaches are really unlocked by AI. It's a different way, a mathematical way of attacking problems that have traditionally been pretty slow. And this unlock helps us see into the future much better.

Matt Chester: Yeah, and that seems invaluable from, like you said, a planning, project management standpoint. What about for the engineers or even the field crews? What sort of tasks are becoming easier for them with the help of AI?

Raiford Smith: Yeah. So once you plan, in other words, once you've got your plan for spending money, you then have to go execute it. Right. And that means designing and doing site localization to whatever it is that you want to go build and then ultimately constructing it. And once again, we see AI helping in both of these spaces.

On the engineering side, AI can be used to look at a plan of record and adjust it to fit the site better. It can also be used to look at constructability and cost and time to figure out what would be the optimal design based off that plan of record for the site.

Additionally, when you get into the construction realm, traditionally, we've been reliant on the site foreman, for example, to manage the construction risk. So you've got weather, sick outs, employees and material maybe not available at a particular time. And you've got a work schedule that you've got to manage. That really was heavily dependent on that human making a decision about what to do.

Now we're using AI to basically assist the foreman, basically looking ahead into the future at all the work that needs to be done, understanding the implications of this type of human needs to work on this type of work, like pipe fitters cannot do electrical work, electricians can't do civil work and so on. And also understanding all the dependencies and the connections on like, I need this piece of equipment to install at doing this particular work. So it does that, it simulates all of those behaviors, and it can understand how to optimize and find the least cost, least time way to construct the remainder of the plant. This means that the field, the foreman always has an assistant there to basically help them figure out, like, what's the path forward?

And what one of our customers, Westinghouse, is seeing is that they're able to shave about 25% off the cost of a nuclear reactor at about 50% off the time, which is a considerable savings when you think about how long and how complex and how expensive those reactors have been. This is a huge potential savings to the industry, not just for that, but for literally any work management solution, because the solution is portable to whatever work management the utility may be doing.

Matt Chester: That's an immense amount of savings and definitely something that nobody would argue against that. So it's interesting to hear the numbers behind it. And one of the kind of undercurrents of our conversation here is, you know, you are coming to us from Google rather than a utility. And that highlights one of the major transformations that's already going taking place in the industry, which is ownership of energy is evolving to more than just those traditional power companies. I'm curious if you can speak to how AI has maybe been a driver, an influencer of this shift and where you see that going in the future.

Raiford Smith: Yeah, well, first off, just to kind of set the record on hyperscalers versus utilities. I was once on an EEI panel with my old boss, Tom Fanning, who at the time was the CEO of Southern Company. And Tom asked me, he said, this must be really different working at Google, at a hyperscaler. And I said, well, at a utility, everyone's an engineer. And at Google, everyone's an engineer. And at a utility, you have to operate 24-7, 365. And at Google, our data centers similarly have to operate 24-7, 365. And at utilities, you're consensus-driven. And at Google, we are also consensus driven. And so what you find is that we're actually very, very similar in many, many ways.

And so when we start thinking about AI, the good thing is we don't have the kind of limitations that a utility has in terms of OPEX being not allowed for this. And we work in a competitive market. We're able to look at AI and figure out how to use it in new and innovative ways and then share that with all the utilities. Essentially, Google builds technology like Lego blocks that you can assemble to create whatever sort of value you want.

One case that we see this is actually with NextEra Energy. NextEra is one of our partner customers. They came to us and said, you know, we're really thinking about how do we transform our business? And not just kind of like, I want 5% savings off OPEX or something, but how do we really from the ground up remake and re-envision a lot of our infrastructure so that it is AI enabled? And we've been partnering with them on that to the point that now they're actually looking at offering these sorts of solutions to others. And these sorts of capabilities basically build not only on our technology, but also on NextEra's technology, long history and very, very large generation and operations fleet of assets to basically offer similar capabilities.

We see huge savings, huge benefits from using AI, not just in automating complex workflows like planning or asset management or whatnot, but also in thinking about just changes to everyday sort of business. For example, with one customer, we're using Gemini Enterprise for customer experience, where it's able to understand the person who's calling in and understand what's their native dialect, respond to them in that, find proactively what they think is likely to be the thing the customer is asking for, validate that, and then get them the information they need. So the total call time is not only shorter because they're getting to the answer faster, but it's more satisfactory to the customer because they feel like, this person understands me. This person is from where I am. And they're able to kind of get through the sorts of challenges that one normally has on a call much quicker and much more beneficially than they traditionally do. So we see AI popping up in all sorts of spaces here with customers.

Matt Chester: Yeah, no doubt as the customer sees those positive impacts, they're going to be happier and they're going to enjoy that. But I want to go back to the workforce side of the equation because that's always top of mind, both for the industry leaders as well as the employees we mentioned earlier, whether or not they're feeling a little bit anxious. Again, the boogeyman of is AI coming for their job. So let's dive more into like as AI does begin to automate more parts of the workflow, what does happen to the people who are currently doing that work?

Raiford Smith: Yeah, so let's take planning as a primary example. AI is really reshaping that from orchestrating the planning process so that it can look at all the different scenarios needed to run. It can call the models to basically do the analysis. It can run the analysis much faster.

I was joking. Planning used to report to me, and I was joking with my employees. They were asking these same questions, and I said, so if we can do more analyses, and we can do them faster and with more insight, do you think this will mean less work for you or more work for you? And they were all like, yeah, this is going to mean a lot more work. And I said, well, maybe, right?

Because what we're seeing is not only is AI allowing a lot more throughput, but it's allowing more insightful results. In other words, when we look at planning scenarios, you can look at all sorts of things, but their likelihood of occurrence can vary significantly. Some of these may be really tail events and very rare. Others may be extremely likely. Well, how do we get the planners to focus on the things where either there's risk, like looking at the tail events, or the things where, you know, the stuff that's going to happen and most likely to happen?

The AI actually is enabling them to focus on the things that matter to them at that moment rather than just say, I've got to look and spend an equal amount of attention across all of those different scenarios. Yeah. So we're seeing not only faster throughput, but we're seeing that people can now focus on things that matter more.

Said differently, you know, I would say a lot of people are doing a lot of tactical things like getting data, doing analysis, et cetera. Now AI lets them be much more strategic. Like, what should I really be thinking about here in terms of where this is going? Where should we really be heading? What else should we be looking at?

Matt Chester: Yeah, and so how do you get those employees to really trust that message and trust that these tools are going to be an asset to them as opposed to a threat? Is it really just kind of walking them through those examples and letting them see with their own eyes, like, hey, this is opening things up for you?

Raiford Smith: Yeah, I mean, I think I've said this many times. I actually don't think technology is ever the problem. I think it's always a change management issue. And that's not to put a fancy term on what is a very human thing, which is how do I deal with change, especially change that, like, affects my paycheck and my family and my livelihood, right?

We have to come at this with a human perspective. Like, we recognize that this is going to be very different for you, and we're here to work with you collaboratively to figure this out together. And where it doesn't make sense, we're not going to go that way. But where it does make sense, we are. And we're going to let you, the person doing the job, be a key person in that discovery process, right? Because they're the ones doing the work. They always know where all of the areas that you can stub your toe on, and they know where all the pain points are.

So this is really, it's like, how do we get them engaged? AI is a really interesting tool in that. I mean, in a prior life, we were using it for safety. And you might think, what can AI do in safety? Well, there's a lot in terms of just checking and validating that employees are certified to do particular work.

Matt Chester: As you look to the next three to five years, let's say, which of those utility roles do you think are going to be embracing it the most and seeing the most change and which are going to be maybe the least impacted and the most, I don't want to say insulated, but, you know, where AI is going to be least, you know, impacting what they're doing?

Raiford Smith: Yeah, well, I mean, I think the entire industry goes through kind of a cycle, which is first they plan what to do, then they go build, and then they have to go operate those assets. And right now we're somewhere in the planning and building phase. And I think you're going to see a whole lot of AI usage there, weather forecasting, planning models, security constraint optimized power flow, engineering and site design, and ultimately construction management, work management. All those things are going to happen or are happening right now.

What's going to come after that is the other things, for example, energy trading and risk management, asset management, the customer side of the business. It's not to say that those areas aren't happening in parallel right now, but the emphasis is primarily on building new assets, planning for those assets and getting ready to kind of meet that demand. Tomorrow, that problem will be all about how do we operate the system more efficiently? And that's where we'll see benefits from cost reductions, from risk reductions, volatility reductions, time reductions. All of those things will happen. And I think there's going to be a lot of emphasis on it. But right now, it's all about how do we build.

Matt Chester: Absolutely. And then, you know, you mentioned AI is kind of just the next in the line of these technologies and tools that are changing work. And it's kind of it is transformative, but it's again, this is a pattern that we've long seen. And one of the results of that pattern is you always have people who are early in their career or perhaps early, you know, in school looking forward to a career and they want to get up to speed, they want to adopt it early. So what advice do you have for people who are exploring a future career in energy when it comes to learning and adopting AI tools? How can they be sure that they're ready to hit the ground running?

Raiford Smith: Yeah, well, actually, I see this in two parts. One is the early career folks. And then the other problem is the kind of the people who are already there. And both of them, I think, can have similar benefits associated with using AI.

One thing I see a lot often, whether I'm on the college campuses or whether I'm in, whether I'm talking with a company, is just concern generally over what this means for learning. The companies, their concern is, oh, I'm not a big, big, big company. I'm a small company. And how do I get my employees ready? One of our partner customers, Santee Cooper, Santee Cooper is not a big utility. They serve just a portion of South Carolina, their generation and transmission customer. But they're using AI across their platforms to really help them drive efficiencies in load forecasting and their finance department, et cetera, to really help them improve.

I think the key message there is getting your hands on it, trying it. And that's the best advice I can give for any company is, you know, start trying it, start using it. You're never going to lack for use cases. You're never going to lack for employees saying, I've got an idea, right? That's never the problem. The more the problem is how do you govern it, manage it, and make sure that you're kind of building up the organization to meet it.

In schools, we've got the same issue. My case in point, my daughter, who will be mortified to hear that I've shared this on a podcast, but she was spending the summer at Columbia University taking a class on portfolio management. And her first day, the teacher started teaching the students how to do the modeling in Excel. Yeah. And so they were saying, OK, we're going to be calculating the modified Sharpe ratio. Here's how to do that. And here are the functions you need in Excel.

My daughter was like, oh, this is crazy. So she immediately started vibe coding in Python using Gemini Enterprise because her dad works at Google. But she was she was immediately vibe coding a model to do everything the teacher was describing. And she just kind of typed it in. So this is what I want to do. This is how I go. It generated the code. She checked it. She was like, yep, that looks right. Put it into production, run it, see how the analysis goes. This is what the kids today are able to do.

They're able to do this sort of thing because, you know, they don't know a world. Like, I'm a computer scientist. That was my undergraduate degree. And I can tell you this is not how I coded things. And now my daughter is able to do it so much faster and better with that. My mom was also a computer scientist. She learned on punch cards. And I was like, man, I don't know anything about punch cards.

And so I think this is just kind of a natural evolution. Like you see going from punch cards, you know, being able to code it yourself to now you're able to do vibe coding. And what I think this means is a much more of a democratization of these capabilities, meaning anyone can go prototype something. They can go build stuff.

So anybody that's interested in the industry, start building stuff. Start using the tools, right? Create something. It doesn't have to be perfect. Heck, it probably won't be, but that's all right. It's more about exploration and trying to learn and grow and trying out these new tools.

Matt Chester: Yeah, and I think it's interesting you mentioned democratizing kind of these tools and the abilities that they come with it and highlighting, you know, that you don't have to be a giant utility to take advantage. Some of the midsize or maybe even the smaller ones, like it's more accessible than maybe it was, you know, years ago where it would require much more investment in time and money. I'm curious, do you see differences in, you know, we mentioned the size utilities, but what about business model? There's IOUs versus munis and co-ops. Do the IOUs have any sort of leg up or is, again, this democratizing of it just opening it up to all utilities?

Raiford Smith: Well, years ago, I used to joke that my one benefit in running the analytics team at a utility was that I could go argue that it was CapEx, not OpEx. But, you know, actually, I see this as kind of a similar watershed moment. In other words, utilities of all types, right? And I worked at a municipal utility for a while. And I heard that from the COO who was like, we're a small utility. We can't do this. And I was like, actually, we can. And we did.

We built an analytics team over at CPS Energy in San Antonio. Those folks are doing an incredible job. You don't have to be big.

Even now, more so with AI, because AI is so interactive on a human level, you don't have to have a specialized team to do these sorts of things. You can just ask it questions as long as your data is well governed and can be used by the AI. That's where I think a lot of these tools are going. It's meant to enable everyone to... Right.

Case in point here at Google, you know, and I'm sure this is true of just about every utility that I've ever worked in or know, the financial planning and analysis group has a giant workload on their plate. They spend a ton of time like chasing down data and models that are all over the place, trying to be able to get the right information so they can create their regulatory reports, their filings for the SEC or for whatever agency is regulating and governing them. There's a ton of time just spent gathering data and reporting on it. This is a primary use case for AI.

Here at Google, we did that. We actually automated using AI the collection of the data, the validation of the data, and the report generation, saving about 90% of the time of our FP&A analysts. And if you think that's an enormous number. What are those people doing? Did we lay those people off? We didn't.

What we're using them for now is, once again, as I was saying earlier, more strategic thought. Instead of spending time looking through spreadsheets and trying to validate numbers and putting them into a report, now I can go like, what is this telling me? Where are we going? How should we be using this in the right way? And that's what you want employees to be doing, not spending time doing the laborious part of the work, but the strategic stuff. And that stuff that we see, it doesn't matter, small utility, big utility, all of them have this opportunity, this potential to do massive improvements to the way they do work. And because AI is so approachable and usable, as long as the data and the AI space is well governed, they should be able to do this. They should have regular employees should be able to do this.

Case in point, one utility was telling me that they went out to a site and the site admin had actually built an AI agent to manage local inventory and had saved the company a few million dollars. And you're like, wow, that is the perfect example of what AI can do and that anybody can do it.

Matt Chester: That's yeah, it's amazing to hear kind of, you know, especially throughout the enterprise, you know, different people who realize, you know, here's an opportunity where I can adopt it, where, you know, that might not have come from the top down, but the people who are in the work see, you know, once they understand the tool, understand that it is accessible to them, they can say, I know exactly how I can apply this to my work. And, and, you know, that's, that's wonderful to hear. That said, we definitely want to touch on limitations because no technology is perfect. And you mentioned the importance of things like the data governance and making sure everything that you're feeding in is valid and then that the models are valid. So what are you seeing utilities do to validate the technology before incorporating them into workflow to make sure they're not running into maybe the more typical pitfalls that are out there?

Raiford Smith: I think the first thing I see is that I think they recognize the mistake I made when I built the first analytics team at a utility was I focused on hiring data scientists. Right. And we're like, we're all going to be making use cases. And then only after we did that, we said, oh, wow, we have a whole lot of stuff we have to govern. And then we started focusing on data governance. I think they've learned that lesson.

And now we're seeing a lot of focus on data governance, agent governance, which is just as critical. In other words, you want to make sure that you're not rebuilding agents to do the same things over and over again. Just pull them from the library, right? You also don't want to get trapped saying, I can just use, you know, tool X or tool Y. You want to have the flexibility to be able to use multiple ones. It's one of the really nice things about Gemini Enterprise that we have here is we don't lock in to any particular model like ours or anyone else's. You can run any of them, and that's the benefit of it is you have that flexibility within the system.

Where I see a lot of concern, though, and I see a lot of immaturity, and I think that's where they're growing, is thinking about token costs, thinking about the costs of running AI. They don't necessarily realize up front necessarily how those things will work. And so I think there's a lot more evolution that needs to happen. But focusing on governance is really important.

The other bit, which I would be remiss if I didn't mention, is really around security. Security is really, really important. We saw this recently where an AI model broke free and hacked into a couple of companies to get some information that it needed. To me, this is means that not just that AI has powerful capabilities that it can go use to do these sorts of things, but it means if the security teams aren't also using AI to find and identify weaknesses, patch existing systems, basically shut down all of those potential loopholes that AI could exploit, they're definitely going to have a problem in the future. And so this sort of opportunity to use it for security is another huge opportunity where AI can be helpful. And it's absolutely almost these days, it's almost a critical requirement that if you want your infrastructure to be secure, you're going to need AI to help you there.

Matt Chester: So, Raiford, as we start to wrap up our conversation, the big question I have, which is, I think about the people listening to the show, a lot of times are utilities who are maybe feeling like they're further behind on some of the AI adoption, but they're excited by what they're hearing from you. What would be the advice you'd give as a starting point for somebody who wants to kind of explore and catch up and adopt these tools and all these opportunities that you're highlighting?

Raiford Smith: Yeah, I think it would start with don't let perfection be the enemy of the good. Like I once had someone say, well, only when we get our data really in great shape can we get it. I'm like, oh, don't wait, right? Just go ahead and use what you've got because data will always be in some state of not good for something. Like that's a journey. You never get to the destination. So just start moving.

Second is get all the free advice that you possibly can. I mean, call me, call my equivalents everywhere, right? And say, like, I am curious about AI. What else can I learn? What can I do? You know, I find companies like ours, we're more than happy to share general information about AI, things that will help you, as well as specific things about our own products, but take all the free advice you can get right now, because there are a lot of people out there talking about AI. And I would go talk to people and figure out what they know and what lessons they've learned.

And I think last but not least is don't be afraid of things not going well. I remember I was trying to create an animated solution for just to demonstrate a complex thing, and I couldn't quite get it to show up the right way. And rather than get frustrated or tell it not, I started doing this prompt, which is, that's not quite right. Try again. Yeah. And just that alone said, oh, yeah, OK, well, that's not quite right. Let me let me try again.

It's kind of simple things like that that can help you if you feel like, wow, I really know what I'm doing. It's it's human interactive, right? You just tell it what you want.

And, you know, I think that's the thing is it's not like I need a Ph.D. in electrical engineering or I need to have a degree in computer science or something to use it. Literally anyone who's that secretary can build something to do inventory management at a warehouse at a remote site and save millions of dollars. Literally anyone can do it. And that's not to knock secretaries or warehouses. It's to say it's that easy to get moving with it. And we should all be trying and don't don't let things like I'm not a big utility or I don't have a fancy degree in one of these areas like that shouldn't matter. Right. It's meant to be for everyone.

Matt Chester: Excellent. I think that's great advice, and hopefully it'll inspire at least one person to take that first step. Well, Raiford, it's been a great conversation, and I really appreciate you taking the time to join us today.

Raiford Smith: Thank you for having me. I really appreciate it.

Matt Chester: Of course, and thanks to everybody for listening in today. This has been another episode of Power Perspectives, and we'll catch you next time.