I can assure you that I have looked into the code in question and discussed it with developers who have years of experience with this functionality and particular code path. It's true that threads are starting up. I agree that it does not seem logical to start up threads, especially during exit, if they don't need to do any work, but that is how the internal logic has been crafted. I would not call these "Enlighen threads" but rather worker threads which can potentially do Enlighten work, when relevant. It has been verified that the crashes we have seen so far with the message in the end of the log had noting to do with Enlighten.

The message is truthful in the sense that it is logged when the threads are starting up. The logic for when to start up these threads is part of internal logic. If these threads actually do some work, which they won't unless Enlighten is needed, other logged messages will inform us about such details. This message has previously led investigations in the wrong direction. When Enlighten is actually relevant, a stack trace and possibly other log messages will provide the relevant information.
In short, we do not say that the message is wrong, just misleading, and that it is not adding relevant information for investigating crashes.