Basically, any scientifically/1st principles viable technology that might save us from doing extensive desperation geo-engineering deserves some investment, from a cost/benefit point of view. Otherwise, we might learn that we coined the name "Antropocene" too early.
There is no viable future scenario in which mankind's energy usage decreases. It's a pipe dream, and ultimately a dangerous one.
Agreed. There would need to be crushing totalitarian rule over everyone on Earth to prevent progress, and that wouldn't be stable long term. Otherwise, competition by competing powers will inevitably result in a Kardashev 2 civilization. Here's why:
Space solar power and fusion power are a couple of serious tech/economic inflection points. Once humanity's technology and our self-organization reach that level of capability, we are literally within striking distance of threatening any star system in our galaxy with huge destructive energies. We only need one of the two for this to be the case. Fusion is unnecessary at this point. If someone can establish large-scale industry on the Moon and the asteroids, solar power will be every bit as good as fusion in many regards. (At large enough scales, solar power is fusion power, basically.)
Modern geopolitics are dominated by the logistics of energy transport and production. It has been this way since before World War II. The logistics of fueling factories, ships, then tanks and airplanes has absolutely dominated strategic thinking since even before that time. The lengths to which the allies went to provide the fuel to run the invasion of Europe were absolute feats of engineering, planning, and intelligence/disinformation. So just think ahead a bit: what if, instead of having to fuel your military, you could instead beam power to outposts, which would use that power to synthesize fuel? The side that can do that would have huge logistical advantages, which would also be strategic advantages. (There's a compelling reason why submarines and aircraft carriers are nuclear powered.)
Going further, once we are at the level of space industry at scale, the Earth becomes just one location in a much larger context, whose available energy dwarfs the available energy on Earth by many orders of magnitude. At those levels of available energy, we could build things like Shkadov thrusters and Nicoll-Dyson beams. At that point, we'd have the hypothetical ability to move stars and build devices that could fry the biospheres off of planets in the bulk of our own galaxy.