I don't think that's true though. There are vital commodities that have been effectively free in some societies (water and salt, for example). Even energy - wind energy is free and some societies used it extensively, and biomass for burning was abundant in some places.
We can argue whether or not it's likely that energy would get to be so cheap in the future, but there doesn't seem to be much basis for his claim that it's economically impossible.
I think the easier arguments are that transformational technologies like 1000x improvements can dramatically extend growth, even if they don't stop it, and that transformational energy technologies that don't rely on the sun would frankly imply starflight. So the timelines might be off by factors of ten, and in such timelines these things don't feel so necessarily absurd.
If you're seriously at the technological stage where you can put computers with 1000x more power than ours running AI and using 1000x as energy dense of technology, expanding beyond the solar system seems pretty approximately the right place to be going.
Just unlikely. And finite. Even a very long time is finite. Perhaps this is where the economist and physicist differ most.
Price is not universal indicator of value, neither being cheap implies that you can't run out of it. Many Indian cities are quickly running out of fresh ground water. The worst thing a city could do is let market determine the price so rich could use it more. Rationing at a less price is the way municipalities are going, which is more ethical way of doing things.
The same could happen for energy or other resources, while they still being cheap.
That wasn’t really his point. His point was that even if energy itself is cheap there’s a floor to how it is valued relative to the total economy for such a vital commodity. If you don’t like the buying it up argument you can also think along the line of delivery cost, capital cost etc. Bottled waters are not free, e.g.