>>Humanity is going to go extinct People aren't worried about the climate or the environment changing enough to directly cause humans to go extinct. More like

>>Humanity is going to go extinct People aren't worried about the climate or the environment changing enough to directly cause humans to go extinct. More like climate change (plus our huge population) leading to increased resource contention, leading to increased conflict, leading to a superweapons conflict, ergo it's an existential threat. > Nuclear is extremely expensive and prone to going wrong. This has been debunked in this thread and countless others. It's a solvable problem, but we in the US lack the will. > To grab that carbon you need to spend at least as much energy as you got from the combustion. Eh. See solutions involving olivine. It's best to not release it in the first place. Which is why... >>We need to use electric cars >...this solves nothing. ... is bupkis. The fewer ICEs releasing carbon, the better. EVs can be powered by renewables. ICEs cannot (economically) be powered by renewables. EtOH is trash, biobutanol and Fischer-Tropsch gasoline are too expensive and likely will stay so. Addendum on nuclear since it keeps coming up. Five tips from HN's favorite lightning rod mogul would dramatically improve the nuclear industry: 1. Make your requirements less dumb. How much design and regulatory cruft and holdover do we have from the beginning of the atomic age? Not suggesting "move fast and break things" but I am suggesting iterate at the small scale and never stop trying to improve safety, cost, and speed, in that order. 2. Try to delete processes and steps. Smaller, more modular reactors. Use fewer parts, and more off the shelf for what you need. 3. Simplify. 4. Accelerate cycle time. This goes hand-in-hand with Simplify. 5. Try to automate it. This is least essential, and doesn't mean just robots, but more jigwork and assembly line, vs nonrecurring engineering, the better.