Would you rather drink one beer every evening for a year, or drink 365 beers in one evening? All of nuclear power 'safety' regulation is built on the idea that, when the same question comes to radiation, there's no difference. But the empirical evidence is overwhelming that the body repairs small amounts of damage caused by ionising radiation in the same way it repairs damage from other small harms. If this is correct it would mean that the "unsafe at any dose" philosophy that has underpinned the regulation of nuclear energy since the 1970s is wrong. It is premature to conclude that yet: we need more evidence. But if that further research does confirm what the existing evidence suggests, then nuclear power is much, much more expensive than it needs to be, even in places like South Korea that build cheaply, and nuclear power could be cheaper than any other form of electricity. We could go much further than, say, the Fingleton Review or the mainstream proposals for reform that are on the table in the United States. We could have a new atomic age in which the "too cheap to meter" dream was realised.