How to counter manipulation via automated attacks

[jaccurate-cmd]

Ex. (bots, spam, coordinated efforts by individual actors)

While reading through this concept, the concern over the title subject grew stronger. Advances in machine learning has allowed unprecedented levels of manipulation of social media and by extension, people. If this proposed system is to triumph, it must be resistant to these sorts of attacks. Any proposed solutions? I'm currently trying some experiments using the provided formula to find ways of breaking the algorithm as a simulation of these sorts of attacks.