There was a talk [0] at Chaos Communication Congress 2 years ago about the early "political bots hype". Sadly in german and I'm not sure how great the provided translation is. It e.g. calls out "more than 50 tweets per day" being used as criteria to identify bots and concludes with more/better research is needed. Would love to hear an updated version or something alike, if someone has recommendations.

[0] https://media.ccc.de/v/34c3-9268-social_bots_fake_news_und_f...