Is Twitter really full of spam bots?

Elon Musk recently put his $44 billion Twitter takeover on hold because he wants further information on the exact number of fake accounts and spam bots on the global microblogging platform.

While some reports suggest that Musk's temporary hold is nothing more than a sly bargaining strategy by the world's richest man, it does bring up a curious question:

How many spam bots are actually there on Twitter?

A discrepancy in the numbers 

In a public filing on May 2, Twitter claimed that less than 5% of its 229 million daily active users are false or spam accounts. This report, made via an internal review of a selected sample of Twitter accounts, looked at users who are targeted with advertising. 

However, there has been no clear indication of exactly what consisted in that sample of accounts, how many parody or joke accounts were taken into consideration or just what is termed as "false or spam" according to Twitter.

Furthermore, since Musk's recent announcement, there have been numerous reports refuting the 5% false account claim by Twitter, suggesting that the number should be much higher - at least three times as much as what Twitter claims.

What do unofficial sources think?

According to a report by Reuters, researchers believe that at least 9% of all Twitter accounts are bots. However, that number stemmed from a 2017 study, and it is expected that the number should have risen higher in recent years. 

Dan Brahmy, CEO of the Israeli tech company Cyabara, reportedly believes that Twitter has downplayed the number of false accounts. According to Reuters, Cyabara uses machine learning to identify fake accounts and they estimate that the actual percentage of false accounts on Twitter should be 13.7%. 

How does Facebook compare?

Moving over to a different platform, in their fourth-quarter report of 2021, Meta stated that fake profiles constitute about 5% of all active monthly Facebook users. 

In the same report, Meta estimated that 11% of all Facebook accounts are duplicate accounts - which means one user controls more than one account at the same time.

One person opening multiple accounts is a common practice in social media platforms such as Facebook, Instagram and Twitter. 

Read more

Read Twitter CEO Agrawal's email to its staff about its changes

What is Twitter doing?

Twitter has not been quiet about spam prevention.

Reports state that in 2018, Twitter acquired Smyte, a software company that specialises in preventing online scams, spam and harassment. After the acquisition, Twitter removed several accounts they termed "spammy and suspicious".

This caused a massive drop of 1 million active users in July 2018, with a stock fall to follow. 

However, Insider reports claim that Twitter employees are not well-versed in handling false accounts and spam bots, as the detection of such accounts is not a practice Twitter is still too proficient in.

An anonymous source reported to Reuters that Twitter utilises a range of metrics and definitions to understand its own idea of "false and spam accounts", but such measures also limit the accuracy of the detection. The continuous creation of new fake accounts further damages the accuracy of such results, claims the source. 

Before Musk completes his acquisition of Twitter, will he be able to find the true number of false accounts and spam bots on the popular social media platform? Only time will tell.