We have already accomplished a great deal. Compile-time C++ is already fully free of UB, which means a huge chunk of real-world C++ is already UB-free today.

This statement is so categorically ridiculous that I have no choice but to assume that Herb Sutter is deliberately trying to mislead people. I would be very surprised if even 5% of "real-world C++" occurred at compile time. In fact, I would be very surprised if even 5% of "real-world C++" was valid constexpr that just ran at runtime.