1 article on this topic.
A 2,015-participant experiment with a replica Israeli military AI targeting system found algorithmic aversion, not automation bias — but adding explainability features erased that skepticism entirely.