You are right, you can absolutely cause memory leaks with Rc, I haven't considered that. However that kind of raises an interesting point - Rc<T>::clone clearly mutates something under the hood as the reference counter is shared, however its not mut. Having something in the core language that breaks the rules and causes actual problems in practice is kind of an ill omen.

And I meant data races, I have stated in my previous post, that race conditions due to issues existing outside of the language, like external libraries or network requests are not the fault of Rust. Neither are logic mistakes a programmer makes.

Interior mutability is explicitly provided for in the language, and consistently marked under the hood via UnsafeCell<>. Thus, the compiler can always check whether interior mutability is possible, including wrt. reference counts. In practice, the &mut modifier has more to do with the absence of aliasing (much like *restrict in C) than mutability itself.