The Five Convergences (Part IV of VI): AI as Optimizer – AI’s Quiet Revolution

In this episode of AIxEnergy, host Michael Vincent continues the deep-dive series on The Five Convergences, a framework mapping how artificial intelligence is reshaping the electric grid. Episode four explores one of the most transformative but often invisible roles of AI: AI as Optimizer.

Vincent is joined by Brandon N. Owens, founder of AIxEnergy.io and author of The Five Convergences of AI and Energy and Artificial Intelligence and US Electricity Demand: Trends and Outlook to 2040. Together, they examine how AI is not always about steering the system in real time but often about acting as a behind-the-scenes analyst, scanning oceans of data to reveal insights that make the grid smarter, more resilient, and more efficient.

The lesson, Owens concludes, is that AI as Optimizer is often invisible but enormously consequential. Its fingerprints are on everything from fewer outages and faster storm recovery to smarter customer programs and more efficient planning cycles. McKinsey has estimated that predictive maintenance alone could save the global power sector tens of billions of dollars. Multiply that across inspections, storm response, trading, and customer engagement, and the impact is staggering.

Importantly, optimizers and controllers can work hand in hand. Optimizers forecast issues and recommend solutions, while controllers carry out real-time responses. This pairing could become the architecture of the future grid—a layered system where cloud-based AI performs deep analytics and edge-based AI executes split-second decisions.

As Owens puts it, AI as Optimizer is the strategist and diagnostician of the 21st-century grid. It doesn’t seek the spotlight, but by revealing hidden patterns and guiding better decisions, it makes the energy system safer, more reliable, and more user-friendly. Knowledge is power, and AI is now amplifying knowledge itself.

Transcript

Michael: Welcome to A-I-x-Energy, the podcast where we explore the rising intersection of artificial intelligence and the systems that power our world.

I'm your host Michael Vincent, and today we continue our deep-dive series into The Five Convergences — a framework that maps how artificial intelligence, or A-I, is reshaping electric infrastructure from the inside out. This is episode four of six on the topic, and today we begin our next deep dive--this one into the concept of A-I as Optimizer.

Our guest today is Brandon N. Owens — founder of A-I-x-Energy dot I-O and the author of not one, but two reports: The Five Convergences of A-I and Energy and Artificial Intelligence and US Electricity Demand: Trends and Outlook to Two Thousand Forty. Together, these reports form the intellectual foundation for understanding A-I’s physical footprint on the American electric grid.

Today we’re focusing on one of the most important shifts happening in the energy industry — something called AI as Optimizer. Can you explain what that means?

Brandon: Not all AI in energy is about steering or real-time control. A vast and impactful role is simply to analyze—spotting trends in data that help us maintain assets, plan operations, or even interact with customers. Think of it as the smart analyst sitting next to human operators. It processes huge amounts of information and provides insights that make the system safer, more reliable, and more efficient.

Michael: One area you highlight is maintenance. With such a massive grid, how does AI actually make a difference?

Brandon: Well look, the U.S. grid is enormous: over 600,000 miles of transmission lines and millions of miles of distribution lines, not to mention transformers, switches, and hardware. Maintaining all that is labor-intensive and expensive.

AI changes the game through predictive maintenance. By feeding in years of equipment data—things like load levels, oil temperatures, and vibration readings—machine learning can predict which components are most likely to fail and when. For example, a utility can scan transformer health indices and flag units that show subtle signs of breakdown. Replacing those before they fail avoids outages and even fires. Some utilities, like Duke Energy and Exelon, have launched AI-based asset programs aimed at cutting unplanned transformer outages and extending asset life. Industry reports suggest these approaches can reduce transformer failures by 20 to 30 percent.

Michael: Inspection has always been a challenge. How is AI changing that?

Brandon: Traditionally, crews went out in trucks or even helicopters to inspect lines. It was slow, costly, and sometimes still missed small defects. Now, drones equipped with high-resolution cameras and LiDAR gather thousands of images. AI-powered vision systems scan them for cracked insulators, frayed wires, or overgrown vegetation that could cause wildfires.

Michael: Outages are always front-page news. How is AI improving response there?

Brandon: Outage management is another big win. AI models can predict storm impacts by learning from past weather and damage data. If a hurricane is approaching, the model can forecast the number of outages, the likely regions, and even which substations are most at risk. That allows utilities to pre-stage crews, move equipment, and even shut down sections to prevent damage. U.S. labs like Pacific Northwest and Oak Ridge are working on these tools. Some utilities report up to 40 percent improvement in crew deployment efficiency thanks to AI’s predictions. Once the storm passes, AI also helps optimize restoration—figuring out the fastest way to get the most customers back online, which is a complex puzzle that algorithms handle well.

Michael: Optimization isn’t just about hardware, is it? How does it affect customers?

Brandon: That’s right. AI is also reshaping customer engagement. Retail energy providers use it to analyze smart meter data and tailor rate plans. AI can cluster customers by usage patterns and suggest who might benefit from demand response or time-of-use rates. On the household side, apps use AI to break down whole-home electricity data into appliance-level estimates—showing when the fridge is failing or an air conditioner is running inefficiently. And now we’re seeing utilities test large language model–based advisors. Imagine a chatbot that can explain why your bill spiked, pulling together your usage history, weather, and tariff data. It might say, “Your use rose 20 percent during a July heat wave, pushing you into a higher tier, which increased your bill by $30. Consider switching to a time-of-use plan to save money.” That kind of personalized, real-time advice is hard for human reps to match at scale.

Michael: What about on the planning and market side?

Brandon: AI is being used for resource planning, microgrid design, and even tariff modeling. For a microgrid, AI can simulate thousands of combinations of solar, batteries, generators, and demand strategies, then recommend the cheapest, most reliable, and cleanest option. When it comes to rate design, AI can model how customers will respond to new tariffs—like EV charging discounts or solar incentives—helping regulators craft policies that balance fairness, reliability, and sustainability. Even energy trading is being optimized. AI predicts fuel prices, helps decide when to hedge purchases, and supports day-ahead scheduling of power plants to minimize costs.

Michael: And AI isn’t just working on the front lines, right? It’s also helping behind the scenes.

Brandon: Exactly. Utilities are drowning in paperwork—interconnection requests, environmental studies, regulatory filings. AI text processing is being used to scan and summarize these documents, freeing staff for higher-level analysis. For instance, the Pacific Northwest National Lab developed a PermitAI prototype that digested millions of words from past environmental impact statements. It now allows analysts to search and find precedents quickly, cutting out days of manual review.

Michael: So when you step back, what’s the bigger lesson about AI as optimizer?

Brandon: The key point is that AI as optimizer is often invisible. It doesn’t “run” the grid directly, but its fingerprints are everywhere—in fewer outages, lower maintenance costs, faster storm recovery, smarter customer programs, and quicker planning cycles. McKinsey once estimated that predictive maintenance alone could save the global power sector tens of billions of dollars. Multiply that across inspections, storm response, customer engagement, and planning, and the impact is enormous. Importantly, optimizers and controllers can also work together. The optimizer might forecast an upcoming issue, and the controller executes a fix. That partnership could become the architecture of the smart grid—cloud-based AI doing heavy analytics, and edge-based AI managing split-second responses.

Michael: Brandon, this has been fascinating. Not all AI is flashy, but optimization sounds like the quiet force making the grid smarter, safer, and more user-friendly.

Brandon: Exactly. Knowledge is power in the energy world. And AI, as an optimizer, augments knowledge. By revealing hidden patterns and guiding better decisions, it helps utilities and customers alike. It’s the strategist and diagnostician of the 21st century grid.

Michael: Visit A-I-X-Energy. The “Five Convergences” report goes deep into AI as Controller with more examples and governance recommendations.

Thanks for joining us, Brandon. And thank you for listening to A-I-X-Energy. If you enjoyed today’s episode, share it with a colleague, subscribe, and join us next time as we explore the convergences shaping our energy future.