Thanks for this. I was paraphrasing (badly, it seems). The video actually says it better:

To write code that lives in an embedded environment, it has to run in this mode in Rust called “no standard” (#![no_std]) and this mode called “no main” (#![no_main]). Basically you have no access to any of the core utilities in Rust, you have to write a lot of them yourself.

He then explains how embedded code necessarily has global mutability which is “the antithesis” of Rust development.

So yeah, you could make all of those wrappers, but at the end of the day you’ll end up with about the same amount of “unsafe” code as you would making the same thing in C++.

Edit: but if what you said still applies, it does seem like Rust would watch your back somewhat better than C++ would in that it wouldn’t even compile unsafe operations outside of unsafe blocks, unlike C++ to the best of my knowledge where you kind of have to review the code yourself to make sure it only uses the appropriate wrappers.