Wie Facebook das Problem der Fake Accounts löst
Trump wütet, Putin schmunzelt, Zuckerberg verniedlicht: Was Recruiter über Fake Accounts wissen müssen
Ein Gastbeitrag von Anna Kucirkova
Täuschungen waren schon immer ein probates Mittel zum heiligen Zweck. Das Chamäleon in der Natur, das trojanische Pferd in der Antike, die gefälschten Geldscheine der organisierten Kriminalität, der gefälschte Schachautomat des Türken, die journalistischen „Glanzlichter“ der Hitler-Tagebücher beim Stern oder die erfundenen Geschichten des Claas Relotius beim SPIEGEL: Die Liste ist lang – und noch lange nicht zu Ende.
Der Cambridge Analytics Skandal bei Facebook hat auf eindrucksvolle Weise bewiesen, wie das soziale Medium Facebook als Mittel zur Beeinflussung von demokratischen Wahlen missbraucht wird.
Auch im Recruiting ist es an der Tagesordnung, dass Bewerber bei ihren Lebensläufen schummeln. Zeiten der Arbeitslosigkeit werden verschwiegen und Lücken in der Vita beweisen eine nicht ganz geglückte Selbstdarstellung der Bewerber.
Bewerberprofile wandern zusehends in die Selbstdarstellungen bei LinkedIn, Facebook, Xing & Co und ersetzen zusehends wichtige Elemente der Bewerberunterlagen. Recruiter haben die Pflicht, solche Fälschungen zu erkennen – wollen sie teure Fehleinstellungen vermeiden.
Die Bloggerin Anna Kucirkova hat sich in einem bei IQS Directory veröffentlichten Bericht mit der Frage auseinandergesetzt, wie Facebook mit Fake Accounts umgeht. Hier ist ihr Gastbeitrag.
How Facebook Is Solving The Fake Accounts Problem
Are you a Facebook user?
If you are, then you can count yourself amongst a global community that includes roughly 2.2 billion other monthly active users.
That’s pretty wild to think about especially when considering that is active users per month.
However, an even more mind-boggling number is the amount of fake accounts the social media giant deleted in the first quarter of 2018: 583 million.
Yep, more than half a billion Facebook accounts, all fake, all deleted. In the just the first quarter of this year.
To put that number in perspective, consider the top six social networking apps in the US total over 577 million active monthly users. That includes the US users of Facebook, Instagram, Facebook Messenger, Twitter, Pinterest, and Snapchat. Combined.
Some reports are showing that as recently as September, Facebook’s deletion number had topped the 1 billion mark, but it’s unclear precisely what that estimate truly entails.
Regardless, it would appear that the powers that be at Facebook are doing a whole lot of deleting, and very little Facebooking.
So what’s going on? What are fake Facebook accounts and why are there so many of them?
And, what, if anything can we do about them?
What is a Fake Facebook Account?
Back in the good old days, which is basically five years ago, a fake account on Facebook usually meant someone was looking to steal your identity or information. Or someone wanting to cause harm to your reputation or the reputation of someone you know.
While the con-artists and generally unpleasant people still exist, fake Facebook accounts have taken on a far more nefarious aim over the past two years.
Prior to the 2016 election, fake accounts, ads, and posts began populating Facebook in unprecedented numbers. The majority of accounts have been tied back to coordinated sources in Russia in an effort to create division among voters and steer public opinion.
Though all social platforms were targets in the 2016 presidential campaign, Facebook faced the brunt of the efforts to increase the tensions in an already volatile election cycle.
Many of the accounts took aim at already hot-button topics such as gun control, but the majority were merely created to propagate disinformation prior to the election.
Other false accounts, however, move beyond the contentious arena of politics. Fake Facebook accounts often pop up in droves after significant events such as school shootings, spewing misinformation and sowing seeds of discontent when unity should be the prevailing mood.
In one such instance, a picture of Emma Gonzalez, a survivor of the Parkland, FL shooting, that originally showed her tearing a shooting range target in half, was altered to show her instead ripping up the constitution.
The most recent example of the continued use of false accounts came at the end of July when Facebook uncovered an effort focused on upending the midterm elections.
During the latter part of the summer, Facebook noted it had removed a number of accounts linked to both Russian and Iranian sources for posing “inauthentic behavior” or proved “misleading.”
Leading into the 2018 midterms, the activity once again spiked, with Facebook working to regain control before election day on November 6. Facebook axed 32 fake accounts, and though they did not immediately link them to the Russian IRA, they did note that the characteristics were similar to the Russian accounts from 2016.
Facebook is not alone in this, however.
YouTube, the popular social video platform, boasts an upload rate of 400 hours of video each minute. Even with a bustling team of moderators nixing rogue accounts, there is an incredible volume of content to sort through.
Twitter recently suspended over 275 accounts that reflected links back to Iran. However, their problems stem from the fact many user accounts are anonymous. Nonetheless, they are getting rid of close to 1 million accounts per day, with approximately 70 million suspended in May and June alone.
To their credit, Twitter is taking action at the potential risk of negating some of their growth.
No doubt the challenges faced by Twitter and YouTube are steep, but Facebook remains the primary focus of hackers, trolls, and those that hope to stop both. It’s massive base of legitimate users, and its extensive global reach practically ensures this.
Were You a Target of the Fake Facebook Campaigns?
In a word, yes.
If you possessed a US-based Facebook account, there is little doubt you and the other 170 million users in the country at some point fell victim to the false Russian ads and accounts that permeated Facebook.
Numerous reports show that between 2015 and 2017, that Russia’s Internet Research Agency doled out over $100,000 to purchase Facebook ads.
The ads themselves targeted all manner of individuals that showed interest in a number of topics. The most popular included independence, motherhood, patriotism, racial equality, and social justice.
Beyond the ads, which totaled more than 3,500 over the course of the two-year span, the fake Russian accounts also tallied well above 80,000 organic posts on Facebook and more 120,000 on Instagram. When the final impact was measured, its estimated that almost 150 million Americans saw some level of Russian disinformation through those two platforms alone.
What was the Point?
Ultimately, the goal was to poke and prod an already sharply divided electorate and create even deeper divisions.
And it worked.
While plenty of the ads and accounts targeted political topics or candidates, most did not. One prime example includes competing ads – one for a pro-Beyonce rally and one for an anti-Beyonce rally – at the same place, date, and time. The apparent goal being citizens shown putting their difference on display.
A lot of the time, the ads were nonsensical or riddled with errors so extensive it would be clear a non-native English speaker cobbled it together. Yet, the bait worked time and again thanks in large part to the overall tone of the ads appealing directly to people’s emotions, with little concern to which side those emotions fell on.
What is Being Done to Stop Them?
As evidenced by the account deletions earlier in 2018, Facebook is working overtime put out the fire of false accounts.
In May, Facebook instituted new parameters to combat the problem. The extensive list of checks and balances include quite a few requirements meant to clearly identify political ads as such and only through advertisers who have been authorized to post them.
In addition, Facebook will archive all political ads (or any other advertisement that Facebook considers of national importance) for up to seven years. Page admins and those who oversee ad accounts will have to provide Facebook a copy of their ID to prove a valid mailing address (with a letter and access code sent to the admin for use on their account).
In one last step, the disclosure of who paid for the ad will also be necessary.
More than just guidelines though, Facebook is also gearing up its physical review process.
It announced plans to increase the number of individuals who oversee and moderate the platform’s content to over 20,000 by the end of 2018. This effectively doubles the amount they had at the end of 2017.
On top of those efforts, the social platform continues to delete fake accounts as soon as they go active with millions blocked every day.
Although many of these safeguards are in place for the 2018 midterms, the real test comes in two years during the next presidential election. By that time, hackers and trolls would have enough time to circumvent any safeguards, assuming Facebook makes no further changes.
Even with the security improvements in place, there are no absolute guarantees. Facebook’s leader, Mark Zuckerberg, agrees, noting in a Facebook post back in April that they would be unable to put an end to “all people trying to game the system.”
Investment analyst firm Stone Fox Capital highlights some numbers how Mark Zuckerberg deals with the challenge of content security and fake accounts in an analysis published by SeekingAlpha.
Remember that Facebook promised hiring thousands of people to review posts for abuse, while Twitter (TWTR) has regularly discussed automated solutions without hiring tons of content reviewers. The expectation here is that Facebook starts moving to more automated solutions such as AI to reduce costs and improve the results of blocking abusive content by removing such content before users even see it.
At the end of December, the social networking company had 30,000 people in the safety group with an incredible 15,000 content reviewers. Note that the company only had 33,600 employees at the end of Q3.
During the Q3 earnings call, CEO Mark Zuckerberg made mention of updating systems and tools to combat abuse and misinformation on the platform:
Our systems for proactively identifying harmful content are improving. Our systems for detecting interference in elections are a lot more mature now… We’ve reduced the incentives to spread misinformation. We’re partnering more closely with governments and outside experts to improve security, including here in the U.S.
The company didn’t aggressively talk about such plans as one should expect in 2019. In such regard, the CFO continued to cut expense growth outlooks in a likely move to not tip off regulators about the intent to move away from reliance on reviewers.
As the problems with fake Facebook accounts continue to persist, it’s up to individual users to be diligent and avoid falling victim to the false accounts. Admittedly, it’s not always easy.
It also shows that we remain susceptible to the issues that plagued the 2016 elections and if no significant changes actually do occur, we will be in for a repeat in 2020.
Alex Stamos, Facebook’s former Chief Security Officer, agrees. He states, “If there’s no foreign interference during the midterms, it’s not because we did a great job. It’s because our adversaries decided to show a little forbearance, which is unfortunate. In most cases, throwing an election one way or another is going to be very difficult for a foreign adversary, but throwing any election into chaos is totally doable right now. That’s where we haven’t moved forwards.”
Of course, even without the rampant use of fake accounts and bad actors, Facebook has long been a platform used for political discourse. However, for any users that want to steer clear of the false accounts, it may require keeping a wide berth from any and all politically motivated content.
While that indeed may be challenging for some in today’s highly charged atmosphere, for others it may prove a welcome reprieve from the vitriol.