Our team was in Houston last month for the 2026 Energy Drone and Robotics Summit. Four conference sessions, with four major energy providers (Duke Energy, Southern Company, Entergy and PG&E), and by the fourth one the pattern was too consistent to ignore.
The pilots are done. Nobody in that room was still proving drones work. They all operate are mature programs now, not pilots. Entergy alone is projecting 500 terabytes of inspection data by 2027, and told the room almost exactly what it wants: a way to integrate GIS mapping, metadata and workflow tools into prioritisation heat maps, so pilots can go back to pure flying instead of chasing data. Southern Company was in the same session, describing the same shift the rest of the room was living through. The conversation has moved past capture. It has landed on a harder question: what happens to all of it, and who is accountable for what it finds.
Three things stood out, and each one points to a different job that needs doing.
Most vendor conversations still start with ROI on the drone itself. Duke Energy's framing was different. They are looking at experienced line workers retiring over the next five to ten years, and they are treating AI as a way to hold the institutional knowledge that walks out the door with each one. That is not a cost conversation. It is a continuity one.
The answer is not faster capture. It is systems that can hold judgment that used to live in one person's head, and make it available to the person who replaces them.
Duke described running redundant surveys because nobody fully trusted the first pass. PG&E has split piloting from analysis into two separate functions entirely. Entergy's own ask, GIS-linked prioritisation instead of raw footage, is the clearest version of the same problem. Every operator in that room wants their people out of the handling and into the judgment. Nobody wants more footage. Everybody wants fewer hours between a camera seeing something and a person deciding what to do about it.
That gap, between image and decision, is where most of the value is sitting unused right now. Closing it is an architecture problem, not a data volume problem. More storage does not fix it. A shorter path from detection to a decision a person can act on does.
FAA Part 108 shifts compliance responsibility onto the operator and opens the door to autonomous operations at scale. It means the providers who can prove governance around what their systems detect and decide will be in a different competitive position to those who can only demonstrate detection.
This is the one I would put in front of any operator right now: start documenting your model governance before the rule requires you to. Being ready before it is mandatory is worth more than being compliant after.
Drone and sensor technology has matured. None of this is any longer about who has the faster drone or the better sensor. It is about who can close the loop from image, to decision, to action, and prove that chain on request when a regulator or an auditor asks for it.
What I am confident about is this: no operator solves it alone, and no single tool solves it for them. A drone that captures beautifully, paired with a platform that has never been tested against your actual asset base and your actual failure modes, gets you nowhere near the answer. Any system is only as good as the data underneath it, and data quality is not something one piece of hardware or software guarantees on day one. It gets earned over years, in the field, against real edge cases, by whoever has been doing the unglamorous work of getting it right.
That is why this has become a partnership question rather than a procurement one. The operators actually making progress are not shopping for the single best capability on a spec sheet at the lowest price. They are choosing to build a working relationship with providers who have already been solving for exactly this, in exactly this kind of environment, long enough to have the governance and the data discipline sorted before the regulation makes it mandatory.
That is the important conversation we keep having with our clients in the harshest in-field environments.
Hanno Blankenstein is CEO and Co-founder of Unleash live, an enterprise computer vision platform. He has spent over a decade deploying AI-based visual intelligence across energy, mining, utilities and critical infrastructure.