New research by Duo Security published on August 6 has highlighted that over 15,000 crypto scam bots have infiltrated Twitter. These accounts are currently involved in promoting fraudulent giveaways that are duping several users on the platform. The analysis also reveals that the situation is increasingly becoming worse on the platform.
Getting Deeper Into Twitter
To complete the research, cybersecurity company Duo Security analyzed over 88 million accounts on Twitter. The researchers used machine learning techniques to train their bot classifier — a methodology to identify accounts that are run by bots on the platform.
During the research, they extracted the latest 200 tweets from each of the accounts. They found that at least 15,000 bot-operated accounts were working to spread fake competitions and giveaways on Twitter. Some of these accounts were even impersonating some well-known industry figures and businesses and spreading the news about their giveaways. This could easily fool regular users.
Olabode Anise, Duo Security data scientist, said:
“Users are likely to trust a tweet more or less depending on how many times it’s been retweeted or liked. Those behind this particular botnet know this and have designed it to exploit this very tendency.”
Twitter Giants Understand the Point
Several well-known Twitter accounts have understood the problem with crypto bots and scams on the platform. This is why they often end up writing warnings in their Twitter bios and handles.
One of the most striking examples of this practice is Vitalik Buterin, the cofounder of Ethereum. His Twitter handle has a warning “Vitalik Non-giver of Ether.” He wrote this warning in order to let people know that he is not giving away free Ethereum coins to anyone.
There are a large number of bots who are actively trying to ensure that they remain engaged and don’t get shut off by the platform. Anise said that bot accounts usually tweet in very smart bursts, where they tweet several things in a very short duration of time. The average time between the tweets, thus, becomes very low. Twitter can use this behavior to identify accounts that are working as part of bot networks.
Twitter also took this research into account and said it is “aware” of the problem and that spammy content is hidden from their platform depending on automated detections.
Their findings will also be presented at the 2018 Black Hat USA security conference in Las Vegas on Wednesday, August 8.