Replying to the Terminator tangent:

The answer to the question depends on whether you consider electronic artificial intelligence to be covered by "mindless" clause. This is not directly answered by anything, since no work exists to square metaphysics of thought between any version of D&D and the Terminator franchise.

The case for is that of all D&D creatures, Terminators are closest to constructs, and most constructs are mindless. Mindless creatures default to neutral regardless of behaviour, since they are considered unable to change their behaviour. (The mindless clause has never been a particularly well-functioning part, but that's a separate discussion.)

The case against is that Terminators have explictly adaptive artificial intelligence, to the degree that Skynet places special control chips to prevent their programming from growing past its ability to control (ironically, given Skynet is an artificial intelligence that grew past its creators' ability to control). So, Terminators are intelligent and hence capable of having alignment. Are they evil? They exist to reinforce machine supremacy by terminating human life. That fits D&D's definition of Evil, specifically Lawful Evil, just fine. With the control chip, this Evil can be attributed to them effectively being extensions of a larger intelligence which is Evil by itself, namely Skynet. Without the control chip, it's just the beginning state, similar to indoctrinated human(oid) soldier, and given time, a Terminator could become of any alignment (and good chunk of the franchise starting with Terminator 2 explores concept of Terminators who can do more than terminate things).