Energy policy makers tout nuclear power as a major tool to transition the world away from fossil fuels, while environmentalists have an almost knee-jerk reaction against them, dating back to the "Nuclear Power? No thanks!" slogan of the 1970s and 1980s.
Some data center operators are now actively hoping that nuclear power might become available for microgrid solutions, to make their facilities independent of the grid
The truth is that nuclear reactors can generate electricity with a much smaller environmental footprint than fossil fuels, and arguable create less emissions even than solar power, when manufacturing costs are taken into account. However, after decades of development, actual nuclear plants are too cumbersome and slow to deliver, when the world needs to get off oil and gas right now.
And efforts to develop a lighter, alternative nuclear industry will no doubt prove useful in future, but are all going to miss the deadline for decarbonizing our current world.
Nuclear power plants provided 17 percent or more of the world's electricity capacity by 1996, according to nuclear industry reports, but they always faced opposition, and after the Fukushima incident in Japan, new nuclear plants have been thin on the ground, with many countries including Germany, reversing previous policies that relied on nuclear power.
Today, nuclear gives us 10 percent of our electricity worldwide, but most of that comes from ageing plants that are due to be retired shortly.
At the same time, the UN IPCC has said that the world must halve its greenhouse gas emissions by 2030 (compared to 1990 levels), and achieve Net Zero by 2050. This means we need a rapid ramping up of clean energy that does not produce greenhouse gases. The nuclear industry has put forward nuclear generation as a major contributor to this effort, but there are questions about this.
Firstly, our current nuclear power stations aren't going to help. The current generation are being phased out, and adding to the problem by removing clean energy. Any new plants will take ten years to build, taking us beyond 2030. They can be a help, but won't meet the immediate need to halve emissions.
Objectors suggest that the money spent on nuclear plants could be better invested in renewable sources, efforts to reduce energy use by measures like home insulation, and developing battery storage to match renewable power with demands.
Other nuclear options?
The nuclear industy has alternative suggestions. Small modular reactors (SMRs) sound promising, as these can be built more quickly from pre-fabricated parts, according to a single design which can be approved for multiple locations. In the UK, Rolls-Royce is hoping to build some 16 SMRs. However, the build time is still four years, and there is a lot of approval still to get through before any of these are off the ground.
In the US, NuScale is the closest to reality, and its approved reactors in Idaho are due to be finished in 2029 and 2030.
Other SMR proposals include TerraPower, founded by Bill Gates. However, most SMRs require higher concentrations of fissile U-235 in their uranium fuel than earlier nuclear plants - and TerraPower just had to put back its plans by a couple of years, because at present the higher yield "HALEU" fuel can only be bought from Russia.
There's fusion, of course. That's a genuinely exciting option, in which light elements are reacted to create heavier ones, just as happens in the sun, potentially without harmful waste, and wih a plentiful fuel source.
While we all want to see working fusion reactors, with the best will in the world, no one could expect this to realistically help with the urgent near term need to displace fossil fuels.
This week, there was massive excitement that the US DOE-backed National Ignition Facility (NIF) achieved "scientific breakeven", getting more energy out of a tiny pellet of fuel than was put in by focused lasers.
The NIF used "inertial containment", essentially imploding a tiny bead of fuel, and the breakthrough, was genuine, but nowhere near a viable energy generator. The energy powering the lasers was two orders of magnitude greater than the energy the lasers delivered to the target, so NIF is miles away from a real "breakeven".
And the scale of the demonstration only produced about three-quarters of a kWh of energy, so it's way smaller than any useful generating capacity.
Scientists still need to find ways to scale fusion up to bigger energies, and to extend the lifetime of the reactions they create, to make something that looks like a power station. Magnetically confined toroids have been the most favored avenue, as they could potentially burn for long periods. But they have been fabulously difficult to achieve and preserve.
Inertial confinement systems might find ways to make their lasers more efficient, and implode a series of fuel pellets, augmented with multiple layers of fuel.
There's another contender - Z-pinch containment, which uses the magnetic field generated by a current in the plasma to contain the reaction. There's a community working on this around the University of Washington, with a startup, Zap Energy, that has a proposal to convert a coal power station to fusion in the future.
But all of these developments are decades away. They might make things easier in future, but they won't exist till after we have passed the 2050 Net Zero deadline, so we need other technologies to get us to that point.
The people working to make nuclear practical are doing the right thing, and we wish them all the best.
The people presenting this as a viable current solution are either peddling unrealistic idealism... or worse, they may just be pushing some cycnical hype designed to take our minds off the urgency of the situation.