The Grid Doesnt Need More Power Plants It Needs This
read summary →TITLE: The Grid Doesn’t Need More Power Plants (It Needs This) CHANNEL: Undecided with Matt Ferrell URL: https://youtu.be/pLIatO-RA1c
In April 2025, most of Spain and Portugal went dark. A cascading failure knocked out the Iberian Peninsula’s grid in seconds. Just four years earlier, Texas came within 4 minutes and 37 seconds of its own total collapse. Not a temporary blackout, a full shutdown. What engineers call a black start, a process that could take days to weeks to recover from. Not to mention all the people that died as a result. The infrastructure that failed them. Well, according to the Department of Energy, 70% of the US transmission lines are over 25 years old. We’re running 21st century lives on a mid 20th century grid. But here’s the thing. One guy actually saw this coming. Back in 1997, energy consultant Carl Robbago wrote a blueprint for a radically different grid. His model, well, the internet. Seriously, the internet. And no, I’m not talking about today’s internet, which is just five billionaires in a trench coat. I’m talking about the ’90s internet. decentralized, collaborative, and just really, really cool. So, how would the internet stop a blackout? And why did the guy who figured it out get ignored for 30 years? But first, we need to understand what went so wrong. I’m Matt Frell. Welcome to Undecided.
This video is brought to you by Outskill. The United States energy grid is many things. A lifeline, a miracle, a house of cards. But one thing it isn’t is current. Let me rephrase that. It doesn’t reflect our modern world yet. The Department of Energy puts it bluntly. Our electric infrastructure is aging and it’s being pushed to do more than it was originally designed to do. The US grid is split into just three regions: East, West, and well, Texas. More than 60 balancing authorities coordinate energy flowing through 180 million power poles, 5 1/2 million miles of transmission lines, and thousands of power plants. Their job, make sure supply matches demand every second, every day. If that balance breaks for even a moment, components get damaged, infrastructure weakens, or worse, the lights go out. And when a blackout looms, our backup plan is basically turn things off and just hope for the best. One of our last ditch protocols called underfrequency load shedding is just automatically flipping circuit breakers until the grid stabilizes. We’ve been doing it since the 1960s. When that doesn’t work, utilities roll out rotating blackouts, cutting power to alternating groups of customers. When that doesn’t work, it’s a total reset. That’s what nearly happened in Texas. When temperatures dropped in February 2021, electricity demand surged while generators went offline one by one. Urkott, the Texas grid, scrambled to rebalance, but it took days of blackouts to stabilize the grid. They were 4 minutes and 37 seconds from needing that total restart. Blackout or black start. When the grid fails, things get dark. And the strangest part, someone wrote the fix for this in 1997. Not a research paper, a full blueprint. It’s been sitting there for almost 30 years. So, why hasn’t anyone built it?
So, here’s where I’ll get personal for a second. I have over 40 solar panels on my roof. I have 20 kwatt hours of energy storage in my garage. My geothermal heat pump handles heating, cooling, and hot water so efficiently that my HVAC barely registers on my electric bill. When my grid goes down, honestly, I sleep pretty well at night. It’s not a brag. That’s the point. The binary of we have power or we don’t only exists when you’re completely dependent on a monolithic grid with monolithic power generation. And today we have cost-effective ways to break that dependency. Rooftop solar, community solar farms, batteries, EVs that can power your house. These tools give you power in every sense of the word. You can become what the industry calls a proumer, part consumer, part producer. You sell excess power back to the grid. You chip in during times of strain. And most importantly, when things go sideways, you’re not helpless. This applies to residential and commercial buildings. So, what would the world look like if everyone had that kind of choice? What if we could pick our energy the way we pick our internet providers? Well, it turns out there’s a name for this, and the guy who coined it accidentally proved it works, thanks to his own mother. But I’m getting ahead of myself. Before I get into that name, and yes, there is a story about his mom, let me tell you about today’s sponsor.
Just like this story, it’s about doing more with what you already have. Smarter tools, distributed solutions, taking back some control. That’s how I describe what AI has done for running this channel behind the scenes. Not for generating videos, but for the unglamorous stuff. AI has quickly taken over parts of my job that used to eat just time. Importing data, analyzing our production pipeline in Notion, automating the boring stuff so that I can spend more time on the interesting stuff. That kind of AI I’m genuinely a fan of, which is what today’s sponsor is actually about. It’s called Outskill. It’s a free AI education platform and it’s already reached over 10 million people across a 100 countries with a 4.9 out of five on Trustpilot. They’re holding a two-day live AI mastermind this weekend, April 11th and 12th, from 10:00 a.m. to 7:00 p.m. Eastern time both days. It draws experience from over 100 practitioners, including people who’ve built AI systems at NVIDIA and Microsoft. The focus is practical. You’ll learn how to build AI agents that execute tasks for you. Automate workflows that run while you’re not watching and connect tools like Sheets or Notion into systems that actually save you hours every single week. But here’s the coolest part. It’s free. The first 1,000 people get full access at no cost. You’ll also get over $5,000 in bonuses if you attend, including an AI prompt guide and personalized toolkit builder. Just by registering, you’ll get access to their AI survival handbook, too. the links in the description or scan the QR code. And thanks to Outskll and to all of you for supporting the channel. Now, let’s get back to what the name of this new energy future might be.
Part of the beauty of the internet is that there’s no singular plug to pull that collapses a network of networks. Just because one ISP is doing maintenance in New York doesn’t mean that everyone on the East Coast is locked out of their phones. Just because one website is down doesn’t mean every other website is unavailable. The helplessness of an internet outage only gets serious when what you need is controlled by a single entity, which is kind of exactly my point. We saw this play out in back-to-back incidents in the fall of 2025 when centralized internet service failures left millions stranded. Now, chances are that you’ve heard of crapification that’s been plaguing online services. So, let me introduce you to something that doesn’t stink. Internetification. It’s a proposal to take the internet’s most powerful feature, it’s decentralization, and apply it to energy. At its most basic, this is about micro grids. Small self-contained power networks that can connect to the main grid, but also disconnect and run on their own. Think of it like your home Wi-Fi. Your devices all talk to each other inside your house. They can reach the wider internet when they want to, but if your ISP goes down, your local network keeps humming along. So, what makes a micro grid micro? Well, it comes down to deers or distributed energy resources. These are the solar panels, wind turbines, and batteries that changed the direction energy moves. Up until the last few decades, delivering energy was a one-sided affair. A utility generated or bought electricity, sent it to you, and that was that. You get what you get. No choice in where, when, or how. Deers flipped that script. That’s what turned you from a passive consumer into a proumer or someone who both uses and produces energy. But beyond individual empowerment, deers are what make micro grids resilient. When more people, communities, and companies produce and store their own power, there’s less pressure on the grid overall, fewer demand spikes, fewer emergencies. When the main grid goes down, a micro grid can disconnect and keep running on its own, which is what engineers call islanding. That’s why hospitals, grocery stores, and data centers already rely on them. In hospitals, for example, backup generators usually aren’t enough to keep life support equipment like ventilators or diialysis machines running for more than just a few hours. Micro grids in healthcare facilities can work as a safety net that protects patients. Right now, the energy grid is a lot like the internet in terms of markets. Some people are lucky enough to live in areas where they can freely pick their service provider. Others aren’t, like myself. Having multiple ISPs competing in the same area is a good thing. It drives innovation and it doesn’t threaten the overall health of the net. Micro grids extend that same freedom of choice to energy. And we’ve known this works for quite a while. The concept goes back 30 years and it starts with a guy who calls himself a regulatory hitman.
Carl Robbago has been on every side of the energy debate. Texas public utility commissioner, Department of Energy official, vice president of Austin Energy. He spent three decades inside the system and three decades arguing that the system needs to change. In 1997, he wrote a book chapter in the virtual utility. In it, he basically described the grid that we need today. One message of the new information media is that truly revolutionary growth in the ubiquity and use of information came only with decentralization and nearly unfettered interconnection. If this message has an analog in the electric utility industry, it is the distributed utility model. Rather than the brute force solution of adding a new power plant or expensive transmission upgrade, the careful targeting of modular and flexible efficiency or renewable options offers least cost options to enhance service quality and reliability. That was in 1997, the same year most of us were still on dialup.
I sat down with Carl and asked him, “Well, after 30 years, has anything changed your mind?” His answer was blunt. The unescapable conclusion is that the distributed architecture for so many things is the most elegant and efficient from nature, you know, to communities to the electric grid. But Carl didn’t just write papers about it. He tested it and some of his best stories sound less like an energy policy and more like sitcom. At Austin Energy, Carl ran a thermostat control program covering 95,000 homes. One day, he stopped by his mom’s house for lunch. She was cranking the air conditioning before a scheduled demand response event. She’d gotten the notice and decided to pre-cool the house before Carl, her own son, turned off her electricity. Here’s the punch line. She accidentally proved the entire concept. When customers got advanced notice, they pre-cooled their homes on their own. The grid got the same demand reduction. Customers stayed comfortable and the utilities sold just as much electricity. So, everybody won. And Carl’s team took it further. Instead of picking a vendor and deploying one proprietary thermostat, they did something radical. They said, “Bring your own thermostat. You pick the device. Just make sure it connects to our API. We’ll keep a secure system behind the firewall and you’ll see pricing in real time. And most important, if you reduce your consumption, we’ll pay you. I’ve participated in a similar program in Massachusetts. And it was absolutely fantastic. Sound familiar? It’s open standards, customer choice, security at the boundary, innovation at the edge, and that’s the internet model, but applied to energy in Austin over a decade ago. You can check out my full in-depth interview with Carl on my second channel where we’ll go into everything from net metering to what it means to be an efficiency advocate. But the takeaway I want to focus on here is where the idea goes next. Carl had the theory. He proved the concept in Austin, but he was missing one piece. The actual structure or device that could make it all work. It took another 28 years and a guy nicknamed Broadband Jesus to propose the solution.
Just a year after Carl published the virtual utility, a Swedish tech entrepreneur named Jonas Berguson built something that would change the world. In 1998, he created a router-based broadband network that freed Sweden from dialup. It earned him the nickname broadband Jesus. Thanks to him, today’s kids will never know the screech of a modem handshake. Decades later, Bergerson aimed at a bigger target. In September 2025, he and his Swedish American research team published Energyet explained, Internetification of Energy Distribution. The pitch, take the exact model that decentralized the internet and build an energy version. Open-source, not proprietary. Energy routers instead of internet routers. Energy net instead of internet. I think you get the idea. The key to the whole thing is one technical trick. It sounds complicated, but it’s actually really elegant. The key innovation is something called galvanic separation, which is also really fun to say. Galvanic separation. It draws a hard boundary between your local energy and your larger grid. The paper puts it this way. Power flows across the boundary when and only when both sides explicitly agree using a predefined digital negotiation process managed by software. This makes it possible to support the grid while it will never be disturbed. Jurgensson said it more simply in a January interview. If you have photovoltaics, that’s on one port. If there’s a fault, it can’t get past that port. If you have storage, it can’t get past that port. Because we can never create a cascade, suddenly we can have an unlimited amount of electrote can vested locally without any technical negative effect on the traditional grid. When I asked Carl about energy routers, he didn’t just endorse the idea. He said he’s one device away from living it. I have a Ford Lightning. I have a couple of solar panels on the new garage. I have intelligent and controllable devices. And the only thing I’m missing right now is the router which could orchestrate all those pieces for me. So yes, I think the router is the invention of this decade. Then he took it one step further. Imagine your router talks to your neighbor’s router. Sarah next door is throwing a party. You’re away for the weekend. Your excess solar flows into her house. The transaction happens automatically. No utility middleman. Just two routers negotiating on your behalf. That’s not science fiction. That’s an architecture that Burgusen published and Carl has been imagining for 30 years.
If this all sounds too good, it kind of is, at least for right now. Utilities aren’t rushing to hand over the keys. Their entire business model is built on centralized generation and delivery. And Carl told me they’ll fight this like they fight dealership requirements for Tesla. Anything to protect the incumbent. And even for early adopters, the transition is messy. Carl described it as a confusing array of options. Time of use rates, demand response programs, solar incentives, smart panels, EV charging schedules, and that’s just one house. He’s a self-described energy nerd, and even he has trouble keeping track. Heck, I’m trying to live this right now in my house, and I find it very confusing. There’s also the access problem. Not everyone can afford solar panels, batteries, or an EV. The proumer revolution means nothing if it only reaches the people with money. For micro grids to work like the internet, they need to be as accessible as the internet. And we’re just not there yet. So, is this actually going to happen? Or is internetification just going to join the long list of great ideas that utilities buried? I’ll tell you what I think, but first, there’s one stat that changed my mind, and it keeps me optimistic. Every time someone claims that a certain level of solar penetration will just break the grid, that threshold keeps climbing. It used to be 5%. Then the national lab said they couldn’t find any real problems below 15%. Now Holy Cross, a real co-op in Colorado, runs on 85% renewable resources. As Carl put it, apparently they’re not all living in caves and burning tar soaked sticks for light at night.
Technology for a smarter, more distributed grid exists right now. Smart panels, solar batteries, EVs are all real, all available. Carl Robbango wrote the blueprint in 1997. Jonas Berguson published the architecture in 2025. We’re 28 years into this idea and we’re just now catching up. I think we’ll get there. Not because utilities will have a sudden change of heart, but because the economics will make resistance pointless. When Carl was at Austin Energy, engineers from the generation side used to sneak over to his team and whisper, “Can I get assigned to your project? I think this is the future.” They weren’t thinking about leaving the utility. They were thinking about what being a utility worker in the future would look like. And that’s the thing that most people miss about this. Internetification isn’t about destroying utilities. It’s about evolving them. We’ll always need someone to maintain the toll road, enforce the speed limits, and collect the fairs. But the road itself, it can look a lot different. As Robo wrote all the way back in 1997, the power of the technological innovation does not lie in doing business as usual in a different even more efficient way. The heartbeat message of the future is fundamental change. But what do you think? Are you team deer or team leave things as they are? Jump in the comments and let me know. You can also check out my extended cuts of many of my videos over on Patreon. If human written and research videos matter to you, Patreon support helps a ton. And speaking of that, a big welcome and thank you to new supporter plus member John M and new producer Therel Fox. The links in the description if you’d like to join, but just watching, subscribing, and turning on notifications is awesome. And check out my follow-up podcast, Still to be Determined, will keep this conversation going. Keep your mind open. Stay curious.