I know this particular post is somewhat old, but the topic is still relevant.
And they're advocating for low CCT which is actually dangerously bad, just like the HPS lights it's replacing. Many studies have been done on visual acuity versus color temperature. Here are a few:
No, a 2700-3000K, 75+ CRI street light is not "just like" a 2200K, 20 CRI HPS street light, nor is it "really, really bad." It's a huge improvement over HPS lights. At worst, ~3000K is only slightly worse than ~5000K, and are far more similar to each other than HPS is to either, but the small differences in acuity and depth perception aren't the only factors at play.
And those studies you linked are only of black and white, static subjects under flat lighting, and at least 10 lux light levels. No glare, no truly low light levels, and no real concern about preserving night vision - to be clear, 10 lux is a minimum goal for a streetlighted area, but the level will be far lower as you leave the lit area.
Proper shielding most certainly does help (and improves efficiency by reflecting light where it is needed), and is part of the AMA recommendation. Yet even well shielded and aimed fixtures still produce some spill and glare. The only way to effectively avoid glare is to increase the lighting angle, which means either significantly increasing pole heights or spacing lights significantly closer together so their beams can be narrowed. Neither option is very practical. Reducing CCT, however, is.
There was unsurprisingly quite a bit of discussion among those involved in street lighting decisions following the AMA policy statement, and one of the ensuing comments that sounded sensible to me was from the city of Phoenix. They indicated their plan was higher CCT lighting along busy streets and in commercial/industrial areas, and lower CCT lighting along less busy residential streets.
The AMA really has no business sticking its nose into a matter like this. Streetlight design is solely the prevue of lighting and traffic engineers.
If only the real world could be so handily isolated to intended consequences. You might as well say the AMA has no business sticking it's nose into air pollution either - that's solely the purview of automotive engineers.
Trying to dismiss blue light melatonin suppression as pseudoscience isn't helping argument. Are you perhaps thinking instead of blue light hazard? There is also real concern about blue light retinal damage, but only at very high intensities (staring directly at a bare, undiffused emitter for an extended period of time is one case I've seen referenced. Welding is another). Concerns about blue light hazards from conventional lights does indeed sound like pseudoscience.
I'm old enough to remember similar complaints back when we switched from mercury vapor in the 1970s to HPS. That's all we have here-the usual complaints from people about something new.
The AMA publishing a policy statement based on a review of multiple avenues of ongoing research is a very different situation from people with no expertise at all complaining that things are different than they're used to.