Software robots, or bots, originating mostly from botnets, are making a huge impact on the internet throughout this entire election season.
Unsurprisingly enough, researchers have dug up evidence of how many social media-based bots are influencing the election.
According to Emilio Ferrara, a computer scientist and researcher, about one-fifth of the entire election conversation on Twitter was generated by a bot.
Ferrara and his co-author Alessandro Bessi, a visiting research assistant at USC’s ISI, in a study, made a discovery indicating that a significant amount of political discussion on Twitter was initiated by pro-Donald Trump and pro-Hillary Clinton bots, with the purpose of distorting the online conversation.
In the study, researchers analyzed 20 million tweets created between September 16 and October 21, all pertaining to the US election.
What researchers found was that bots produced over 3.8 million tweets, or 19 percent of the social conversation on Twitter, as published in a Daily Dot report.
On top of that, researchers also revealed that Twitter-based bots were responsible for 400,000 of the 2.8 million users examined, which translates to almost 15 percent of all the participants in the study.
In the journal First Monday, a paper was released titled, “Social Bots Distort the 2016 U.S. Presidential Election Online Discussion.”
“The presence of these bots can affect the dynamics of the political discussion in three tangible ways,” the paper reads.
“First, influence can be redistributed across suspicious accounts that may be operated with malicious purposes. Second, the political conversation can become further polarized. Third, spreading of misinformation and unverified information can be enhanced.”
As described in the paper, Trump’s tweets that were created by bots had almost entirely self-promotional commentary, profoundly inflating his popularity.
When looking at Clinton’s botted commentary, only half of them were positive, with the other half slamming the nominee, according to the paper’s findings.
It was also noted that South Carolina produced the fakest campaign-related tweets produced purely by bots.
The engineering behind these Twitter bots is complicated. Although it’s used by businesses, politicians, and governments, the “master puppeteers” who create these bots are widely unknown.
It is said that Russian and Indian deceitful, blackhat digital marketers produce these Twitter bots to try and generate some monetary value.
These bots, which are often created on Facebook and Instagram, use identical names of real users, along with similar biographical information, and even profile pictures. The accounts are also able to retweet, share content, comment, and even engage in human-like discussions on Twitter.
Donald Trump’s following on Twitter has been called out for having an inordinate amount of these fake accounts in the past.
Nick Bilton, reporting for Vanity Fair, dived into Trump’s fake Twitter following and said most of his followers at the time of the election were nothing more than bots.
“But if you really want to see fake, you have to look at all the accounts that retweet Trump,” Bilton wrote in his Vanity Fair piece. “If you go back through his timeline, you will see that at around the same time last year, Trump was averaging just a few dozen retweets a day on a typical tweet, and 100 to 200 likes on a tweet.”
As Bilton noted, looking at an account’s sharing ratio can determine whether or not a user is manipulating Twitter’s algorithm to amplify vanity metrics or deviating from the terms of service.
Bilton says that the next time Trump boasts about his Twitter following, don’t be too proud of his success.
“So next time he sends a note saying, “Thank you! 22.4+ million followers!” know that half of the people he is talking about are just figments of Donald Trump’s imagination. Perhaps that is fitting amid an election cycle that only he could have dreamed up.”