Most FSD users are in it because they see how great things could be if self-driving existed, and are willing to put in the risk to get the algorithms trained.
How does that risk differ from human drivers? Other than the fact that the training would affect many vehicles vs just one driver, and the fact that humans are only supervised with possible intervention for 6hrs total.
