Skip to main content

Benjamin Perrin is a law professor at the Peter A. Allard School of Law, University of British Columbia

When a Ku Klux Klan supporter, an online lubricant retailer called Lubezilla, and 1,000 other random Twitter users start spreading nasty negative comments about you, you can rest assured you've been had.

It happened to me on Friday.

This Twitter attack showed me just how terrible automated social media attacks can be: They're used for cyberbullying and to attack political opponents.

Canadians are very active on social media networks: 59 per cent of adults are on Facebook and 25 per cent have a Twitter profile. But, as I learned, not all social media accounts are actually people. So-called "bots" are social media accounts controlled by a computer but appear to be authentic.

Opinion: Outrage Twitter, and the end of the benefit of the doubt

I joined Twitter in June, 2016, to learn more about it for a new research project on crime and social media. It quickly led to debates about electoral reform, Jason Kenney's decision to provincially campaign while collecting an MP's salary, the burkini ban in France and other hot topics.

On Friday, Stephen Harper resigned as MP. By noon, I noticed #GoodRiddanceHarper was trending on Twitter. So I simply tweeted: "#GoodRiddanceHarper is trending in Canada #cdnpoli".

Jason Fekete, a reporter with the Ottawa Citizen, retweeted my tweet, sharing it with his followers, adding a note that I was Mr. Harper's former legal counsel in the Prime Minister's Office. I replied with a link to a CBC story from the closing days of the last federal election campaign where I said Mr. Harper's government had "lost the moral authority to govern."

That's when, unbeknownst to me, someone unleashed the Twitter bots. It began with a nasty tweet about me – fair enough. Within an hour, 100 "people" had shared or liked it. By midafternoon more than 300 had. By dinnertime there were more than 1,000 likes and retweets from haters. I'd gone viral, and in a bad way. Or so I thought.

I decided to click on one of the thousands of "people" who'd joined the online attack on me. It was for a basketball team in Cobb County, Ga. Their sudden interest in Canadian politics seemed odd. Things got even more bizarre.

A Twitter user called TheAmerican, described as a Maryland-based neo-Nazi and supporter of the Ku Klux Klan, retweeted the attack on me. Boutique clothing stores in New York and eventually the aforementioned Lubezilla (which claims to offer "the most extensive and unique range of lubricants in the world") joined in too.

Almost all of the accounts joining this attack on me were based in the U.S. Most had only a small handful of followers, and many had retweeted identical random content.

That's when I remembered hearing about how China uses fake Twitter accounts to spread propaganda and attack political dissidents. The Wall Street Journal has also reported identical, automated tweets supporting Donald Trump were spotted in February.

"Political bots are among the latest, and most unique, technological advances situated at the intersection of politics and digital strategy," says Samuel Woolley who studies them at the University of Washington. Mr. Woolley says political bots have been deployed in China, Russia, Saudi Arabia and Turkey. He can now add Canada to the list.

I began Googling and quickly found a social media bot service which describes itself as follows: "TweetAttacksPro 4 is designed to run thousands of Twitter accounts at the same time 24/7 to auto-follow, un-follow, follow back, tweet, retweet, reply, favourite, delete tweet, unretweet, unfavorite, add to list and send messages to your new followers, etc. Every account can have its own settings, thus preventing Twitter from becoming suspicious about the account, plus the software can simulate human operation perfectly!"

There were not thousands of people suddenly hating me. It was someone in their mother's basement (or work office) using a ham-fisted social media scam to make me think the world was suddenly against me.

I've never laughed so hard.

Twitter prohibits bots and automation of accounts so I reported it. Within minutes, the attack was vapourized (except for the screenshots I kept). I have a pretty thick skin, figured this online attack out and knew what to do to stop it. But imagine if someone had released this unethical and prohibited social media attack on a future Amanda Todd or Rehtaeh Parsons. The devastation caused by an automated cyberbullying attack that can be launched with the click of a button could be tragic.

Lurking in our online communities are bots, who look and act like people – to a point. But they will only get better and more sophisticated. Their deployment in Canada for political attacks is a troubling first.