Most utilities already have more data than they know what to do with. They collect interval reads. They measure power quality. They have visibility into the substation and the major assets above it in the network. So, when we talk about the energy transition, the problem is not a shortage of data. The problem is where the decisions are made. 

That distinction matters more than the industry admits. Fifteen-minute data delivered hours later does not stop a voltage drop, a reverse flow or a local instability event. By the time the information reaches a central system, the moment to act has passed. In a grid dominated by rooftop solar, batteries and electric vehicles, that lag is not an inconvenience. It becomes a major structural risk. You cannot run a dynamic, two-way network with hindsight. 

This is the real shift underway. The grid is not just getting more complex; the complexity is moving. Risk, variability and the moments that determine reliability are all migrating toward the low-voltage edge, closer to the customer. Yet most network investment still points upstream, toward the parts of the system we already understand and can visualise. AEMO forecasts that consumer-owned resources will make up close to 40 percent of the National Electricity Market's capacity by 2050. If that much of the system is going to sit on the customer side of the meter, the question is why so little of our intelligence lives there. 

The Asia Pacific region makes this impossible to ignore. Australia has among the highest Consumer Energy Resource (CER) penetration in the world. Our distributors are already accountable for outcomes at the edge while lacking the real-time visibility and decision capability to influence them. What happens here over the next few years is a preview of what most networks globally will face later. That makes the region a proving ground, not a follower. 

The answer is not to centralise harder or pour more data into the back office. It is to move the decision closer to the event. This is what grid edge intelligence delivers. By processing high-resolution data locally, on the meter, utilities can detect faults earlier, respond to outages faster and manage distributed resources as they behave in real time rather than as they appeared hours ago. This is not theory. Itron has more than 18.3 million distributed intelligence capable meters shipped with 8 million applications in use, running analytics where the value is created. In practice that translates into measurable outcomes. We're seeing greater than 95 percent accuracy in the electrical connectivity model, up to a 50 percent reduction in the effort needed for connectivity and phase assignment, and a potential 10 percent decrease in operational cost. 

To be clear, this is not edge versus central. That framing is close to being a decade out of date. The model that works is local intelligence coordinated enterprise wide. Decisions that need speed, happen at the edge. Decisions that need scale and oversight happen centrally. The two work as one system and the utility keeps its network authority throughout. 

None of this works, though, if the data underneath it is a mess. Our 2025 Resourcefulness Report found that 81 percent of North American utilities already use AI, and 41 percent report it fully integrated, up from 25 percent just a year earlier. Ninety percent of executives rate AI and analytics as critical. But the single biggest barrier they name is not ambition or budget. It is data infrastructure, governance and standardisation. AI is only as good as the data it runs on, and today too much of that data is fragmented across systems that were never designed to talk to each other. A utility operating system that gives standardised access from the edge to the enterprise is what turns edge intelligence from a collection of clever apps into an operating model. 

The same logic applies to distributed energy resources. Orchestrating solar, storage and EVs is not a single leap, it is a progression, and it should be treated as one. It moves from emergency backstop controls to flexible exports, to coordinated dispatch, and finally to autonomous orchestration where the network optimises itself continuously and safely at scale. We see that path unfolding over the following five years. Utilities do not have to rebuild to get there. The most effective strategies build on the metering and communications networks already in the ground, protecting those investments, whilst adding capability in step with business priorities and regulation. That lowers risk and brings value forward. 

The energy transition is not a future scenario. It is happening now, at the edge, faster than our central systems were built to handle. The utilities that lead the next decade will not be the ones with the most data. They will be the ones that move decisions closest to where grid conditions change and coordinate those decisions across the whole network. Everything else is just measurement.