.png)
Civilization is running two clocks at once, and almost nobody is watching both.
The first clock measures energy. Artificial intelligence is hungry. Every model that answers a question, writes an essay, or recognizes a face runs on data centers, and data centers run on electricity, enormous amounts of it. Utilities that spent years planning for flat demand are now being asked for power on a scale they did not design for. Here in New Jersey, where I'm studying toward a career in the energy grid, you can already feel the strain: an aging grid, a growing list of projects waiting to connect, and a public that keeps being told the lights will stay on without being told how. Much of our power still comes from burning fossil fuels, so the more we ask of the grid, the harder it is to keep the climate promises we've made.
The second clock measures people. Across the developed world, birth rates have been below the level needed to replace the population for years. In the United States the fertility rate sits well under replacement. In places like South Korea it has fallen far lower. This isn't a forecast; it's arithmetic that is already underway. The children who were never born will not become the nurses, electricians, line workers, and engineers of the 2040s. A smaller workforce has to carry the same roads, hospitals, and retirees, plus everything we're about to build.
Put those clocks side by side and the problem is plain. We need more energy infrastructure than ever, and we will have fewer human hands to build, run, and repair it.
The twist
Here is the part that makes the story stranger. The same technology that is straining the grid is the one thing that can help us cope with the worker shortage.
AI and robotics are the only tools we have that can stretch a shrinking workforce across an expanding system. An automated inspection drone can survey miles of transmission line that once took a crew days. Software can predict which transformer is about to fail before it does. Robots can handle dangerous, repetitive work in construction, manufacturing, and maintenance. None of this replaces the need for skilled people, but it changes how far each skilled person can reach. One technician with the right machines can do the work of several.
So AI is both the thing creating the emergency and the thing that might let us survive it. That isn't a comfortable position. But it's the honest one.
The race
This is why I call it a race. The question is not whether we will have machine labor. It's whether we get enough of it online before human labor falls below the line where a modern society can still hold together: the line where there are too few workers to maintain the grid, care for the old, and build what comes next.
If machine labor arrives in time, a smaller population can run a bigger, cleaner system, and the decline in births becomes a challenge we manage rather than a collapse we suffer. If it arrives late, we get a gap: aging infrastructure, overstretched workers, blackouts, and rising costs that fall hardest on people with the least. Nobody gets to choose which clock runs slower. We only get to choose how fast we build.
One buildout, two answers
The encouraging part is that we don't need two separate plans. The same energy buildout that answers the climate crisis answers the demographic one.
Clean, abundant, reliable power, from solar, wind, storage, nuclear, and a modernized grid, is what runs the data centers and the robots without wrecking the climate. It is also what makes automation affordable enough to fill the labor gap. Cheap, steady energy is the foundation under both. Build it, and the machines have what they need to work, the climate has a chance, and a smaller workforce is lifted instead of buried.
That means treating the grid as the most important project of this generation. It means faster permitting for transmission and generation, real investment in training the next wave of line workers and technicians, and honest conversations about where the power for AI will come from. It means colleges like mine, and programs that move students from community college into STEM and energy careers, are not side issues. They're part of the plan.
We can argue about almost everything else. But the cliff is real, both clocks are ticking, and there is exactly one road across. The only question is whether we start building it now, or after we've already gone over the edge.