What's messed up about red light cameras is they can actually be useful - if used correctly!
The correct use case is "We seem to have a problem with red light runners at this intersection, so let's find out why by temporarily deploying red light cameras here."
I've seen this done and the city in question found out. They were able to make some changes to the light timing and at several intersections, that caused the amount of red light runners to drastically drop. (It was stuff like the left turn light not turning green when the straight forward light did).
The only experimentation that you need to to is extend the damn yellow lights. Long enough duration of yellow lights reduces accidents to nearly zero. This has been proven over and over 1000x. The data has been out there for 40+ years. There's zero need for red light cameras.
The correct use case is "We seem to have a problem with red light runners at this intersection, so let's find out why by temporarily deploying red light cameras here."
I've seen this done and the city in question found out. They were able to make some changes to the light timing and at several intersections, that caused the amount of red light runners to drastically drop. (It was stuff like the left turn light not turning green when the straight forward light did).