> This why it is imperative that we __turbocharge__ the pace of innovation in nuclear power. TerraPower is just one of dozens of startups around the world that

> This why it is imperative that we __turbocharge__ the pace of innovation in nuclear power. TerraPower is just one of dozens of startups around the world that are now exploring new and better kinds of reactors As promising as nuclear power is as a renewable energy source, the last thing the world needs is nuclear powered by innovative software from VC funded, fast returns oriented, move fast and break things types. Nuclear reactors do not have the same economics as software. The field is a bad investment in comparison (and in comparison to __many__ other fields). Encouraging young, bright people to go into the nuclear power is dishonest. The future for nuclear researchers in uncertain. Change that before encouraging citizens with great potential to gamble their future away. New, much more efficient reactor designs have and will be in the works. Theses designs are in various stages of regulatory approval. Most designs are not economically viable. Rigor is not quick, and is not to be disrupted. Again, change the incentives. (or just have Bill email fellow billionaires who can afford gambles). I agree with the main point. The rest of the article is irresponsible and/or dishonest rubbish that leaves out important details and actionable ideas. Waste of readers' time and doesn't do much to actually educate the population. Maybe this article even encourages people to spread ill thought out and uninformed ideas in the name of the environment. Subpar editing and publishing. --- nitpick below --- I've worked on embedded software in the nuclear industry (designing inspection tools). I've done work with formal methods and verifying programs. Anyone who jabs at "designs drafted during the slide-rule era and adapted from reactors used on aircraft carriers and submarines" is being disingenuous to the level of rigor that went into these PE signed engineering designs. Our modern software development methods are a joke in comparison.