Thanks very much, Stephanie. Welcome to DTEK 2026. It's an absolute pleasure to be on this stage. This conference has its roots all the way back to 1990. Sure, we've had a couple of name changes in many different venues, probably none quite as grand as this. But the thing that allowed this conference to persist for those 36 years is the importance of the electricity grid. There's no more important or perhaps more complex system that has ever been undertaken than what all of you work on every day. And that's not just because of scope or scale, but it's because modern life depends on the grid working perfectly every day. The grid is more than just infrastructure. It safeguards economies. It protects communities and it enables progress for the modern world. And it really is this last point that I want to get into today.
How we manage that transition will define reliability, resilience, affordability, and trust for decades to come. My talk today really centers on that transition, that transformation, not as a slogan, but as a responsibility of all of us.
Change begins with the use of data. Data is foundational to what we all need to do. The grid was originally designed with the notion of one-way flow. Let's build enough generation. Let's just push the power out to the consumption point and all will be well. In that kind of environment, information moves slowly. Decisions could be made based on averages rather than evidence. And that world really no longer exists. Today, we expect the grid to be able to manage wildfires, manage EV charging, distributed generation, aging infrastructure, aging workforces, weather volatility. And we've got to manage all of this at the same time in an error-free environment. That sort of complexity without visibility really means unbounded risk.
The key to managing that is having data. Not just any data, but data that you can trust. And that data can't be a byproduct. It has to be the nervous system of the grid. When you can see local voltage levels, when you can see load at the feeder, you can understand power factors out at every endpoint in the network, you really understand the grid in a much, much different way. You can understand the behaviors across neighborhoods. And when this happens, transformation is possible. The notion of grid edge intelligence that Itron spends a tremendous amount of time talking about, investing in, and deploying with our customers, it really is that very notion of how can we get visibility and control out at the edge.
We've deployed more than 16 million endpoints that are real-time grid sensors and are solving real-world problems as you'll hear from some of the folks that I have on stage in just a couple of minutes. But really bringing the notion of wildfire mitigation to life. When you look at the data at the edge, you can understand where you have a hot tee, where you've got conductor slap, where you do have vegetation interference, and when you can see it, you can take proactive measures to mitigate it. EV charging management, we can bring that type of solution for the good of the grid. How can we make sure that we're not overloading transformers, but we can also take it right down into the home and understand how to solve real-world problems about a level two charger on a dwelling with a 100 amp panel. Outage reduction. When we understand precisely what the condition is, where the power is on, where the power is not on, when we can understand exactly what happened, we can have crews that arrive on the scene knowing what to do, be able to operate safely and get the lights on as quickly as possible.
How about emergency preparation and repair and preparedness? The huge ice storm that traveled all the way across the United States here just in the last two weeks, Winter Storm Heather, left many utilities in its wake, but those that really had the ability to prepare before that event happened were in a much, much better position. So if you can use an AI model to understand where to stage your crew based on the track of the storm, understand what are the assets that are in place at the time, you have the ability to proactively prepare and deal with the crisis in a much more seamless manner.
Digital construction management, the grid needs to be scaled. We have more data centers. We have more onshoring of manufacturing, more electrification of everything. And when that is the case, how do we make sure that we get as-built plans in real time as we're doing the construction and automate that process all the way from the planning stage to the close out of the project. These are just a few examples of ways that data can be put to work and used as the modern language of the grid. Trusted data tells us what's happening, why it's happening, and most importantly, what will happen next. This is not future possibilities. This is present realities.
Of course, there are present realities. Affordability is a real concern and something we all have to grapple with. And we're all starting from a different place. We have different systems, different regulatory environments, different investment paths. Transformation is absolutely not about abandoning what you have built. But it is extending it with intention. Layer by layer, insight by insight, decision by decision. Start from where you are and you can accelerate forward. Technology must be designed to evolve as change needs. The notion of being able to buy a fixed capability and deploy it, advertise it over 10 or 20 years without expecting change in the environment around you, I think that notion is no longer the world that we live in. So how can we make sure we have technology that can be used in the field, grow as we need, modify as the needs around it change and that capability really revolves around the use of the data itself. When we can build momentum that creates confidence and the confidence creates speed which is important in a rapidly changing world.
Here I think the notion of something like distributed intelligence is really transformational. Instead of sending all the data that you have out at the edge back to a central location, you can deploy an agent. You can deploy an application out to the edge of the grid and take action. Meters become more than meters. They're endpoints. They're intelligent grid sensors. They observe, they interpret, they alert, they control out at the edge of the grid. So, you can see local voltage levels, you understand how things are changing, what's happening at every transformer level without actually having to put a sensor on that transformer. All of this is possible with the data that we have. We just have to put it to work.
Immediate outage confirmation, transformer loading, stress visibility, all of these are absolutely possible when you put distributed intelligence to work. This isn't the notion of data collection and it's not even centralization. It is coordination. So it's an entirely different way to think about use of data in the grid because the future will not only happen in the control room. It's going to take the combined knowledge of what's in the control room with the knowledge of what's happening and the control capability out at the edge of the grid. And that knowledge can be understood everywhere with decision-ready capabilities. We move then from a world of reactive to predictive, from centralized control to distributed awareness. And the capabilities here are absolutely software-driven and they will evolve as the grid evolves and modernization becomes practical and not theoretical.
No discussion today is complete without some discussion around AI. If we're talking about technology and AI isn't really the future of the grid because it's already in the grid. It's not going to replace leadership, at least no time soon. It's going to shape how leadership behaves, how it sees, and how it reacts. AI can predict failure. It can forecast demand and power flows, can enhance safety, it can clarify investment. And AI in all of those examples becomes a decision partner and not an authority. And here's the part that I think is very unique to our industry. I think every insight needs to be explainable and every recommendation needs to be traceable. And why that is the case is because we work in an industry where trust is vitally important. And we cannot underestimate the importance of trust in the business. And that trust cannot be built by speed. It is built by understanding and showing up with responsible actions every day. AI absolutely can and will be one of the most powerful tools in the toolbox to help with grid transformation and this is how we have to think about it and how we use it.
That said, no technology alone can transform the grid. It's people that do. It's all of us. The most powerful intelligence in the grid today and for some time is absolutely human. Field crews that show up and know what to do when a storm happens. Engineers that design for resilience. Operators that restore communities when they're out. Leaders that balance supply and demand and risk and responsibility. Sure, the technology supports them, but it's the people that determine the success.
In my career, the most successful projects that I've ever undertaken are not defined solely by technology, but they're designed by the collaboration of those that you're working with. Grid intelligence, distributed intelligence, it only works because people trust it. It only works because we apply it. And AI works because we guide it. That way, transformation is not a product. It really becomes the partnership.
And that really brings me to the conversations that I want to have with a number of key industry leaders. So with that, let me welcome Scott Krider, president of SDG&E, to the stage and tell us what's new at SDG&E.
Yeah. Well, listen, Tom, I think, you know, and this is probably the most exciting time in our company's 140-year history, and I think that could be really said about the utility industry generally right now. So, you know, we're really becoming incubators of innovation, a lot of new technologies, a lot of new customer experiences. And so, I think we're really setting the bar and really moving this industry forward. And so, you know, at SDG&E, we're really focused probably on three core areas right now. And I would say number one, it's modernizing our business. And, you know, as a company that's been around for more than a hundred years, we know that we've got to change. We got to evolve. We got to set ourselves up for the next 20, 30, 40, 50 years for our customers. I think number two, it's around safety and its resiliency. You know, yes, we have beautiful weather here. This happens a lot. But we do get extreme fire risk here. And so, we're continuing to make really, really prudent investments in technology and in grid improvements for wildfire safety as we're also building the grid to handle more EVs. You know, 25% of our customers have rooftop solar today and really getting ready to go to a zero-carbon future down the road. And then finally, and you said it earlier, it's really about affordability. You know, consumers are frustrated with the cost of everything right now. And so we're doing our part to get our cost of service down even more, working with our policy makers to make sure that energy remains affordable for customers.
Very good. We'll get to maybe the technology part of the conversation in a moment, but I want to take a step back. I spent a lot of time just a moment ago talking about the use of data to solve technological problems, but there's also a part of our business which isn't purely engineering-based. So what do you do to increase the use of data or data literacy, if I can use that term, in the non-technical part of your business?
Yeah, Tom, you know, as we think about this, this is probably one of the biggest change management initiatives that we're really going to have. And it's really more than just change. Really we're talking about a culture shift and getting all of our employees from the executive level all the way down to our line-level employees to really begin to embrace technology even more than they've done so, to really embrace the power of data and really getting our employees to really understand that data and technology is no longer just the domain of our IT department or our data scientists or somebody who can code in Splunk or make really, really pretty PowerBI dashboards. This is really about trying to get the technology and the power of data into the hands of all of our employees. And so we spent the last couple of years really focused through our center of excellence doing a lot of proof of concepts, really trying to have kind of a grassroots focus in terms of what's the next thing for us to invest in. And we're going to really take it a step further. What we're actually launching this year is a new AI academy. And it's not just about trying to teach our employees about AI or just the technologies, but it's really that power of the technology and the tools and the data and to get them really thinking differently. And then we're going to couple that with an AI ambassador program to really have peer-to-peer networking, peer-to-peer conversations. Because what we're finding is that the best ideas are not coming from me. The best ideas are not coming from necessarily our technology leaders, but really it's that grassroots down in the organization.
It's awesome. It's awesome. Let's talk a moment on wildfires. Real problem for many utilities, certainly true right here in Southern California. Your wildfire center is absolutely a marvel. So, if folks haven't had the opportunity to see it and you're given tours, I would absolutely sign up for it. But the thing I want to talk about is the use of data and specifically edge data to solve real-world problems and do wildfire mitigation.
Yeah. Yeah. So, you know, Tom, we had a catastrophic wildfire in 2007 caused by our utility equipment and we created an ethos from that moment forward. Our ethos is never again. And so we've been on this journey at SDG&E to make sure that our facilities are not the cause of a catastrophic wildfire. And, you know, we started with one meteorologist and one weather station in our service territory. And really, as we're starting to fast forward, you know, 20 years later, we're really moving from reaction, which is what we used to do. The wind is going to blow, it's going to be dry. What do we do? And now we're moving really into prediction. And so the use of technology and data is helping us make much better decisions, faster decisions in probably a couple key areas. Number one is our AI-enabled weather models. And so we've got about 225 weather stations all throughout our service territory. And so we have hyper-local data now for the past 10 years. And so we've actually trained our AI weather model to have absolute hyper-local prediction. And it's not just it's going to be windy in our mountains, but we know it's going to be windy at 2 a.m. on this particular circuit. So we share that data with all of our first responders, all of our communities, and so we can have a regional response when we know we're going to have an event. I'd say number two is our drone program. You know, really having drones that have taken millions of images of all of our facilities. And when we started this, we had literally dozens, sometimes hundreds of people in a room reviewing these images to see if there's any anomalies on our systems. So now with intelligent image processing, we're able to again really train an AI model to look for broken crossarms, maybe you have a cracked insulator, woodpeckers on the crossarms, you know, it's amazing what you see from above as opposed from being below. So that's really been a game-changer. And then finally for us is our AI camera network. So, we actually started, we're blessed here in San Diego to have some absolutely incredible partners and academic institutions and we actually started with a camera network here in San Diego that's now grown and proliferated and now there's probably thousands of them throughout the West and to where we can actually have now these cameras that are AI-trained to pick up smoke detection. So, we have cameras all over the place. It picks up a smoke detection. It gives lat and long so we know exactly where that smoke is. It's dispatching helicopters and first responders and it's able for us to start running our fire simulator models which we do about 10 million simulations every night with the San Diego supercomputer. So that's really where we're headed in terms of from being a reactive utility to being very proactive with our communities.
Spectacular. The last question before I could spend hours having this conversation but they don't give us that much time. So let's talk about the workforce of the future, the utility of the future. How can you put the use of AI, data, all the things we're talking about to build a workforce for San Diego Gas and Electric for the years to come?
Yeah, you know, I think it starts sometimes it starts at the top, right? So I mean we are making this an emphasis. We're making this a priority and both in our retention of our employees, we are focusing more on training and upskilling and making sure that we want to invest in their future. I mean, right now it can be really scary for people. You know, they hear AI, they hear all these stories about how there's going to be so many job cuts and job losses. And so we're really focused on trying to give some comfort to our employees that this isn't about getting rid of jobs but helping them to make their jobs a lot easier to do and a lot more efficient. And so that's really one area but we're also trying to again trying to show them what the art of the possible is. You know, as we go through and whether it's giving 5,000 of our employees access with the appropriate guardrails and data protection in terms of a large language model and let them actually try this out. And so they're starting to feel a lot more comfortable. And then as we really think about hiring and attracting talent, hey, listen, the utility sector has not always been the place you want to go if you really want to do cool, interesting tech, innovative things, but that is absolutely changing. And so I think between what we're doing on wildfire, what we're doing with data and smart meters and predictive algorithms in terms of usage that there's no better place to be than the utility sector. If you want to have real purpose and a journey and whether it's on tech innovation or whether it's on climate, whether it's on carbon reductions, this really is the place. But it is going to start with people and it's really attracting that talent that's going to be absolutely essential to us.
That's awesome. Thank you so much for your time.
Thank you very much, Tom.
Absolutely. All right, Mike, welcome to the DTEK stage. So, let's start with the notion of how we can help on the affordability crisis. I mean, that absolutely is top of mind for everybody and how can technology be used to help with affordability?
Well, that's great. Thank you, Tom, for having me here. And I'm vice president of strategy and innovation for PG&E and my passion is to use technology to make a meaningful impact on people's lives. And let me give you an example of this. This is something our teams are working on together. So for our customers when they look to electrify, if you go to add an electric vehicle to your home, about half of our customers have a panel at your home that's 150 amps or less and one of the challenges that then faces when you talk to your electrician, they say you need to upgrade your panel. Now all of a sudden it's gone from I just want an EV to I have a home project to do and it could be $5,000, $10,000 to upgrade that panel and then they call PG&E up and sometimes we get to the spot where the service transformer also needs to be upgraded and all of a sudden now it's a $30,000 project there and we have to make a huge upgrade on our side. And so we've been working together. What if we could solve this problem for customers to make it better, faster, cheaper to electrify? And this is where AMI 2.0 comes in for us to use grid edge intelligence to just modulate the load a little bit. It should be unnoticeable for customers to do that and in a way that we can do it to one, save thousands of dollars off of a upgrade at home. Second is save tens of thousands of dollars in utility upgrade costs. And then third is the gift of time here. So we can go from having a project that could take six to 12 months down to pretty much next day. And so it's an exciting thing to be working on in a way that really drives real impact not only in people's lives, but when at scale it's billions of dollars of savings for our customers, which results in real cost and rate savings. It's really important to do right now.
