If we want clean energy that actually works at night, the answer might not be bigger reactors. It might be smaller ones built like airplanes, not cathedrals.
The world runs on too much energy, and most of it is dirty. Fossil fuels heat the planet. Coal plants poison the air. We talk about renewables like solar and wind as if they've already won, but they haven't the sun goes down, the wind drops, and battery technology, while improving fast at companies like BYD and Tesla, isn't yet ready to carry an entire grid through a still winter night.
So what about nuclear? The standard reaction is to flinch. Nuclear means Chernobyl, Fukushima, radiation, waste, and giant concrete silos that take a decade to build and run billions over budget. But that picture is mostly outdated and even where it's accurate, it's describing a specific kind of nuclear that we don't actually have to keep building.
First, let's clear up nuclear's reputation
Nuclear reactors don't pollute in the climate sense. They emit essentially zero COâ‚‚ during operation, which is why a growing number of climate scientists argue we cannot reach net-zero without them. The radiation fear is also overblown living next to a nuclear plant exposes you to less radiation than living next to a coal plant, because coal ash is mildly radioactive and gets dumped in the open.
The real problems with traditional nuclear aren't day-to-day pollution or radiation leaks. They're three different problems: cost, construction time, and waste storage. Western nuclear projects routinely run twice their budget and a decade behind schedule. The Vogtle reactors in Georgia took roughly 15 years and ended up costing over $30 billion. That's the actual reason nuclear has stalled not safety, but economics.
The idea: build smaller, build many
Here's where the interesting shift is happening. Instead of giant custom-built reactors, what if we built small ones on assembly lines the way we build airplanes and shipped them to wherever they're needed?
This is the pitch behind small modular reactors, or SMRs. The logic is compelling:
- Standardized. Same design every time, so safety and monitoring improve with each unit built.
- Faster. Factory assembly cuts construction from a decade to a few years.
- Cheaper at scale. Once you've built the hundredth unit, the cost per reactor drops dramatically.
- Modern controls. Built with current monitoring tech, not 1970s analog systems retrofitted with digital duct tape.
Several companies are racing to make this work: NuScale and X-energy in the US, Rolls-Royce SMR in the UK, TerraPower (Bill Gates's outfit), and Oklo, which is going even smaller with microreactors. China actually beat everyone to the punch and connected the world's first commercial SMR to the grid recently.
The honest catch
SMRs are not a done deal. The "modular" cost advantage only kicks in if you build a lot of them the first few are expensive, sometimes more expensive per megawatt than traditional reactors. NuScale, the most-hyped American SMR company, cancelled its flagship Idaho project in 2023 because costs ballooned before construction even began. Regulatory approval still takes years. Waste is still waste, just smaller amounts of it.
There's also a real argument that the money is better spent on solar and grid-scale batteries, which keep getting cheaper on a remarkably predictable curve, while nuclear historically has not. Battery storage has gotten dramatically cheaper in the last few years and is being deployed at massive scale, especially in China and the US. The "sun doesn't shine at night" objection is less of a dealbreaker than it was even three years ago.
So what's the actual answer?
Probably a mix. SMRs for steady baseload power and industrial heat the things solar and wind genuinely struggle with. Solar, wind, and batteries for the bulk of electricity generation, where they're already winning on cost. And we stop pretending that any single technology is going to solve climate change on its own.
The intuition that we should build small, controlled, iterable reactors instead of giant unmanageable ones is exactly right. It's also harder than the marketing suggests. Both of those things are true at once, and that's fine it just means we have to actually do the work instead of waiting for one perfect solution to arrive.
The future of energy isn't a silver bullet. It's a toolkit.
Top comments (0)