If machines can create significantly more value with the same unit of energy, then the cost of energy will likely go up. Unless energy supply grows equally or faster to meet the growth in demand

If the supply of money increases to match economic growth, then the inflation-adjusted cost of energy might go down. But all indicators seem to be pointing in the opposite direction this decade.

Economic growth is likely going to become bottlenecked by energy, which means higher inflation-adjusted energy prices.

The only way I see it going down in that time frame is massive buildout of power generation coupled with a devastating economic recession, which seems somewhat counterintuitive