Debunked: How the Internet got fooled into thinking Twitter predicted the Paris attacks

The Internet's elaborate theory about what happened 48 hours before Paris. (AFP/File)

A conspiracy theory about the Paris attacks is picking up more steam than the rest, beginning with a single tweet posted 48 hours before the events unfolded Friday night.

Here is the tweet.

People online are wondering how a tweet describing the current death toll in the French capital could have come out two days before the actual killings occurred. 

The answer is a strange mix of incredible coincidence and dumb luck. The account (now suspended) was a bot, automatically creating headlines using the tweets of a real news account called PZFeed Breaking News Feed (@pzf). The bot essentially grabbed parts of old tweets by @pzf and combined them to make fake headlines. These are the two tweets it grabbed to create the conglomerate tweet published on November 11:

BREAKING NEWS: Death toll from Paris terror attack rises to at least 12 - Reuters

— Breaking News Feed (@pzf) January 7, 2015

AFP: Death toll from Nigerian mosque attack rises to at least 120 with 270 others injured http://t.co/fa9Q7aRZHx

— Breaking News Feed (@pzf) November 28, 2014

These are both real headlines published by a legitimate news source. But one refers to a Nigerian mosque attack last November that killed 120 people and injured 270, while the other is about the Charlie Hebdo attack in Paris earlier this year.

It also doesn’t help that the names of both the actual news source and the bot account are so similar—PZFeed Ebooks vs. PZFeed Breaking News Feed.

It could be compared to the age-old question of whether a monkey left alone with a typewriter would eventually write the entire works of Shakespeare.

Previous tweets from the handle show a stream of irrationally combined headlines, just like the one this week. The amount of times the bot used the same death toll with different events should give you an idea of how it works, and how likely it was that one day it would combine both events. (Screenshots via Twitter) 

Most of the world became familiar with the Twitter spam bot with one, glorious account called Horse_Ebooks that sprouted up in 2010. The handle rose to Internet infamy with nonsensical quips seemingly pulled from other sites. Here’s an example.

a man or woman turning their kitchen into an art studio. Life is short let s make the most of it while we are here You will know about

— Horse ebooks (@Horse_ebooks) September 7, 2013

So deeply did Horse_EBooks tickle the millennial heart that when the New Yorker outed its creator as a human—not a bot—Twitter kind of changed forever. It even inspired the birth of many actual bot accounts under the same, EBooks name.

Perhaps that should have tipped people off to the news on PZFeed Ebooks. But it didn’t.

Instead, retweets, screenshots and suspicion abound from fringe and respected news voices alike—including Al Jazeera Arabic host, Faisal al-Qassem—proving that sometimes the biggest coincidences touted by conspiracy theorists on the Internet are actually just that. 

هذا الحساب نشر عن تفجير باريس وعدد القتلى والجرحى قبل الهجوم ب 48 ساعة. ادخلوا على حسابه للتأكد pic.twitter.com/gcBjv93wyA

— فيصل القاسم (@kasimf) November 15, 2015

my newest fear: PZFeed Ebooks

— james (@dark_bIue) November 14, 2015

Look at the Date of this Tweet. #FalseFlag pic.twitter.com/AYRt2SAcWv

— Jamie (@Jamie83107302) November 15, 2015

By Kane Hippisley-Gatherum, Alisa Reznick