Google's AI stoplight program is now calming traffic in a dozen cities worldwide

They should be careful with the term "Traffic Calming" as that is a term that describes using red lights and obstacles to make people stop more frequently and supposedly drive slower through the city. Many cities are doing this, despite the fact that it wastes fuel and time, and doesn't really increase safety. It just frustrates people and makes them drive like maniacs.

This is literally the opposite of Traffic Calming.

I'd like to see Google's solution to distracted driving, but I don't honestly think it's a problem they can solve. Average wait times of vehicles is tied in no small part to how quickly people react to the light turning green. Everyone behind someone who's surfing the web and doesn't move when they should pays the price. At least here in Houston, this has triggered a ton of red light running because (my personal theory) the people selfish enough to force everyone else wait on their noticing a green light are selfish enough to run a red light rather than be personally punished for their not paying attention. Seriously, it's endemic here that you'll see three to five cars running lights, particularly on turns.

"Calming" traffic is presumptive on people following the rules of the road and that's just not happening these days.

I see no mention of the pedestrian experience.

I’m not saying it isn’t considered, just pointing out that there’s no mention in it in the article.

Is it harder to cross the street? Easier?

"Green Light uses machine learning systems to comb through Maps data to calculate the amount of traffic congestion present at a given light, as well as the average wait times of vehicles stopped there. "

Aren’t there analytical solutions to stoplight scheduling? Or do they make inaccurate assumptions about traffic when used in practice?