A typical memory safety issue in a C program is likely to generate an RCE. A thread-safety issue that leads to a segfault can likely only lead to a DoS attack, unpleasant but much less dangerous. A race condition can theoretically lead to more powerful attacks, but triggering it should be much harder.

A thread-safety issue does not always lead to a segfault. Here it did because the address written was 42, but if you somehow manage to obtain the address of some valid value then you could read from that instead, and not cause an immediate segfault.

I agree with the sentiment that data races are generally harder to exploit, but it _is possible_ to do.