SG16, in reviewing the addition of the new characters to the basic character set found a couple of awful ways it could result in conforming code becoming non conforming. 

https://isocpp.org/files/papers/P2558R1.html

#include <stdio.h>

#define STR(x) #x

int main()
{
  printf("%s", STR(\u0024)); // UCN for $ was allowed, now it is not
}

and


#include <stdio.h>

#define EAT(x)

int main()
{
  EAT(\u0024) // UCN for $ was allowed, now it is not
}
I am not sure this was noticed in reviewing N2701 but thought I'd bring it to your attention. 
SG16 had consensus to forward this to EWG.