Elon Musk recently announced that his $44 billion bid to acquire Twitter is “on hold” because of concerns about the number of fake accounts that make up Twitter user statistics. Musk issued a tweet that cited a news article reporting on a Twitter company filing that estimated that “false or spam accounts represented fewer than 5% of its monetizable daily active users during the first quarter.” Musk’s tweet said: “Twitter deal temporarily on hold pending details supporting calculation that spam/fake accounts do indeed represent less than 5% of users.”
Musk’s decision to put a proposed billion-dollar acquisition “on hold” raises a key question: just how many Twitter users are actual, physically existing human beings who might respond to advertising on the social media platform and thus are “monetizable,” and how many are fakes that exist only in a computer, ready to artificially boost tweets and accounts with followings and retweets? And a related, and even more difficult, question is: how do you figure out who is real and who is fake in the Twitter world, where everything is done electronically? Wired has an interesting story about just how tough it is to separate the real from the fake in the Twitterverse, noting that looking at potential indicia of phoniness necessarily involves both subjectivity and uncertainty.
Attempts to quantify the number of Twitter bots out there suggest that there may be a lot of them. For example, Newsweek reported this week on an “audit” of the official White House twitter account for President Biden that concluded that 49.3 percent of his 22.2 million followers are fake, based on analysis of a number of factors that are used to identify bots. Another “audit” of Musk’s own Twitter account determined that more than 70 percent of his 93 million followers are likely fake or spam accounts. My guess is that Musk isn’t bothered at all by that kind of story, because it proves the point that he raised in his decision to put the Twitter deal “on hold” in the first place: there are serious questions about what is real in the Twitter world that should be answered before billions of dollars are paid for what could be an empty, non-“monetizable” gaggle of bots.
I don’t do Twitter or pay much attention to it because the Twitterverse seems like a strange, mean-spirited place that doesn’t bear much relation to real life as I know it. The kinds of “audit” results reported above raise still more questions about the reality of the Twitter world, and whether those raw numbers about Twitter followers and retweets should be viewed with some healthy, human, non-bot skepticism.