[jfindley]

Putting an "AI" camera on a road in Cornwall, UK has no ability to affect how many people are killed on China's roads, and so it's deeply silly to use those numbers to try to justify doing so. There's far less invasive things that could be done to dramatically improve road safety in many of the countries you list.

[bluecalm]

Things like what? Traffic violence and risky behavior if a raising problem and it seems we don't have a solution for it despite many attemps. My view is that most of it comes from a small group of drivers. It's not "everyone speeds and uses their phone sometimes" but "there are people who speed recklessly and are on their phone most of the time". If you share my view then you want the camera in areas with the most traffic as the goal is to fish out and eliminate those dangerous drivers not punish sensible drivers who make mistakes sometimes.

Putting it on a major "safe" motorway is perfect. As you will have a chance to check most cars there. Being and to go at the speed limit where it seems safe (from inside the car anyway) to go faster is also the most important psychological trait for safe driving. Catching people who aren't up for that will improve safety in other places as well.

[mdp2021]

Never said I support the activity.

But let us keep the logic terse. There are dead because of X, and if some practice P reduces X there will be some advocacy to have it adopted around. The numbers are just relative so largely irrelevant to the advocates.

But it is best to have them right so the lower instances summoned by "Millions!" and "One per million!" - out of context if a cost-risk-benefit with quantifications in all parts is not fully construed - can be avoided.