Zombie Media 91
or promoting certain types of news by modifying the algorithms that
presented certain stories in the news feed. As Koene observes,
how visible the site makes the content has a huge impact on how far
it spreads beyond the account of the original contributor. The way
an algorithm works can have a similar effect to a newspaper editor
selecting a piece for the front page.
As such, the filter bubble, by following people’s preferences, makes it
very hard, in the words of Eric Schmidt, for people “to watch or consume something that has not in some sense been tailored for them”.
Pariser has called this the “passing of the torch from human gatekeepers
to algorithmic ones”,
46
but the failure of that algorithmic gatekeeper
became evident in August 2016 when, having fired human content curators, fake news began to trend on the site, with the top story featuring a false headline that Fox News had fired “traitor” Megyn Kelly
for backing Hillary Clinton.
47
Although human editors were reinstated
three days later, the commitment to ensuring that curation takes place
through algorithms remains very clearly at the forefront of Facebook’s
operations. According to the digital marketing site Zephoria, with more
than 1.86 billion active users worldwide and more than half a million
comments per minute, any chance of collating that information through
human intervention is completely impossible. Processing vast quantities of data is something where computers have a clear advantage over
humans.
And yet, as we have already seen, this is only true in terms of a quantitative advantage. In terms of quality, the benefits are far from clear.
Throughout the 2016 election, a plethora of fake news stories began to
trend on Facebook, clearly intended to shape the vote in one direction
or another, as with a false allegation that Barack Obama had signed
an executive order banning the Pledge of Allegiance in schools, or that
Pope Francis had endorsed Trump, or that Trump had claimed previously that he would run as a Republican because they were the “dumbest
group of voters in the country”.
48
As the majority of American adults
now get their news from social media sites, such fake news could easily
proliferate – not merely in the USA but also in European countries such
as the UK, Italy, France, and Germany.
49
Sometimes, fake news stories
have outperformed legitimate stories by the most popular media companies due to the fact that organisations such as Facebook have outsourced
the gatekeeping of such information to end users and software: if the
primary function of an algorithm is to judge a story on its popularity (a
relatively simple and easily quantifiable exercise) as opposed to its verifiability (which, as we have already seen, is almost impossible to quantify), then fake news that appeals to more readers will quickly become
even more popular than unpalatable real reporting.
or promoting certain types of news by modifying the algorithms that
presented certain stories in the news feed. As Koene observes,
how visible the site makes the content has a huge impact on how far
it spreads beyond the account of the original contributor. The way
an algorithm works can have a similar effect to a newspaper editor
selecting a piece for the front page.
As such, the filter bubble, by following people’s preferences, makes it
very hard, in the words of Eric Schmidt, for people “to watch or consume something that has not in some sense been tailored for them”.
Pariser has called this the “passing of the torch from human gatekeepers
to algorithmic ones”,
46
but the failure of that algorithmic gatekeeper
became evident in August 2016 when, having fired human content curators, fake news began to trend on the site, with the top story featuring a false headline that Fox News had fired “traitor” Megyn Kelly
for backing Hillary Clinton.
47
Although human editors were reinstated
three days later, the commitment to ensuring that curation takes place
through algorithms remains very clearly at the forefront of Facebook’s
operations. According to the digital marketing site Zephoria, with more
than 1.86 billion active users worldwide and more than half a million
comments per minute, any chance of collating that information through
human intervention is completely impossible. Processing vast quantities of data is something where computers have a clear advantage over
humans.
And yet, as we have already seen, this is only true in terms of a quantitative advantage. In terms of quality, the benefits are far from clear.
Throughout the 2016 election, a plethora of fake news stories began to
trend on Facebook, clearly intended to shape the vote in one direction
or another, as with a false allegation that Barack Obama had signed
an executive order banning the Pledge of Allegiance in schools, or that
Pope Francis had endorsed Trump, or that Trump had claimed previously that he would run as a Republican because they were the “dumbest
group of voters in the country”.
48
As the majority of American adults
now get their news from social media sites, such fake news could easily
proliferate – not merely in the USA but also in European countries such
as the UK, Italy, France, and Germany.
49
Sometimes, fake news stories
have outperformed legitimate stories by the most popular media companies due to the fact that organisations such as Facebook have outsourced
the gatekeeping of such information to end users and software: if the
primary function of an algorithm is to judge a story on its popularity (a
relatively simple and easily quantifiable exercise) as opposed to its verifiability (which, as we have already seen, is almost impossible to quantify), then fake news that appeals to more readers will quickly become
even more popular than unpalatable real reporting.