All I am going to say about this is this.

There is nothing that better exemplifies how much the C++ has failed in this regard, when you realize that it is getting SIMD before the ability to multiply 2 numbers to get a bigger number.

If that doesn’t shame someone, I don’t know what will.

I think that just tells you that multi-precision arithmetic isn't that useful when you have the abstractions that make use of it.

Java, Python, and JavaScript have all enjoyed some kind of BigInt type for years now, and anything to do with multi-precision arithmetic sits somewhere deep in the implementation, as it should. It's a kind of feature that's useful to a small group of library developers, but not to the general public, and so even in systems languages like C and C++, you don't see it standardized for a long time. That small group of library developers is often fine with using intrinsics.