I'm guessing that the research methods used by the academic groups in the article aren't easily automateable otherwise it would appear that we have the beginnin

I'm guessing that the research methods used by the academic groups in the article aren't easily automateable otherwise it would appear that we have the beginnings of a solution to this type of behavior. First, define the pattern of behavior, then identify instances in the wild, next identify the actor, then name and shame. Throw some legistation behind this with criminal penalities for offenders and you have a regulatory agency that combats misinformation.