The issue with bitcoin is that the amount of energy needed to secure the network isn't actually necessarily well matched with the amount of energy people are in

The issue with bitcoin is that the amount of energy needed to secure the network isn't actually necessarily well matched with the amount of energy people are incentivised to put into it. The amount people put into it depends on the mining rate (which decays as a slow exponential over time), the price of bitcoin, and the price of energy and mining hardware. The amount needed to put into it to secure the network depends on the price of energy/hardware and the amount of money you could make from a 51% attack (which is functionally a double-spend: not only do you need a substantial amount of bitcoin to pull this off, you need to be able to get the cash out somehow. Or you can try shorting it and assume the market will panic when it happens). At the moment the price is high and the mining rate is also quite high, while energy is frequently available below-cost (certainly below environmental cost) due to a combination of factors including corruption and freeloading on subsidies intended for other purposes. At the moment I think the energy use of bitcoin is far higher than is necessary for the network to be safe from a 51% attack, but also in the long run (because the transaction rate is so low and the mining rate dropping over the next few decades), I think mining will not be enough to secure the network, and either it fails due to a 51% attack or the transaction fees become truly ludicrous. I think cryptocurrencies are a good innovation as a whole and it's likely something like them will be used to provide great value in the future, but bitcoin is most definitely not that, and it's position as the first, most valuable, while also the slowest and least efficient makes it not only harmful to the crypto sector by strangling others but also just bad for humanity as a whole. The sooner we can move on from it the better (etherium might be a good option in the longer run: moving to proof of stake removes the big energy waste and they actually have some scope to evolve the system, unlike bitcoin which basically got fossilised in a half-complete state).