From the examples he's mentioned, it sounds like there's quite a bit of [bike shedding] (https://en.wikipedia.org/wiki/Law_of_triviality) in the process for RFC's in Rust. That's a problem when you've got lots of smart people trying to help.
Because it's not often apparent where useful innovations will come from. If we could predict in advance what will be useful and what won't, we would surely save a lot of money and effort but life isn't that easy.
Thing is, ultimately, some of these technologies may yield useful innovations.
NFT's are the biggest load of speculative garbage to come down the pike in years but in terms of smart contracts (which may have some use) they're not total garbage.
AI is in the "Peak of Inflated Expectations" portion of the hype cycle but that doesn't mean it will never yield anything of value.
The Tech Industry isn't really unique in our chasing the latest shiny object behavior. Anyone recall when 3-D TV's were going to be the next big thing? The consumer electronics industry was practically salivating at selling us new TV's every couple of years.