Original article can be found here (source): Artificial Intelligence on Medium
Flatten the curve — how social media platforms are stopping the spread of fake news
Fake news has been a major topic of conversation for quite some time now, and the issue has come up again with the emergence of coronavirus.
“Have you heard we’re going into lockdown?”
“The army is preparing for deployment!”
“WE NEED MORE TOILET PAPER!!!”
In Ireland, Taoiseach Leo Varadkar pleaded with the public to stop sharing unverified information:
“These messages are scaring and confusing people and causing real damage.”
But Leo, who are these verified sources? (He did tell us, this is just for a thing I’m going to do later…)
If you’ve already read Daniel’s post, “Struggles of the YouTube Algorithm”, you’ll know that YouTube has implemented a banner directing users searching for coronavirus to either the website of the World Health Organisation (WHO) or their own national health authority — in Ireland that’s the HSE. Facebook, Instagram, and Twitter have also introduced this measure in an effort to promote the spread of information from verified sources.
** Aside **
So who are these verified sources?
WHO are these verified sources. It’s been staring us in the face this whole time! (That was the thing we talked about earlier.)
(If you haven’t read Daniel’s post yet and would like to, I’ll link it in the references section below.)
** End of aside **
That’s a nice thought given the times, but what about the general policies of companies like WhatsApp, Facebook, Twitter, and Instagram when it comes to stopping the spread of fake news? As you may be aware, three of the four platforms listed above (Facebook, Instagram, and WhatsApp) are now owned by Facebook Inc. According to Tessa Lyons (2018), a product manager at Facebook,
“Misinformation is bad for our community and bad for our business”.
The company claims to fight the spread of misinformation by giving people more context on the stories they see, reducing the distribution of false news, and removing any accounts or content in violation of their policies.
At present, Facebook Inc partners with third-party fact-checking organisations to identify fake news and stop it spreading on their platforms. This works as follows:
- Technology and user feedback is used to identify potentially false stories.
- These stories are reviewed by the independent fact-checkers.
- If a story is rated false, users are made aware and the link is demoted in users’ News Feed.
- Action is taken against repeat offenders, which may include reduction in the page’s overall distribution or removal of their rights to make money or advertise on the platform.
While it’s clear that Facebook Inc are taking action against the spread of false information on Facebook and Instagram, their approach seems very limited in its ability to review all potentially false stories — they just don’t have the man power!
Involving technologies like machine learning and artificial intelligence could be of huge benefit to the company, increasing effectiveness in catching fake news and reducing their current reliance on third-party fact-checkers.
One of the company’s main concerns is the fine line between misinformation and satire or opinion. It can sometimes be difficult to differentiate between intentional misinformation and controversial opinions or satire. On the other hand, questions have been raised over the objectivity of fact-checkers and the process of challenging their decision. Perhaps technology could provide a more unbiased rating.
Where WhatsApp is concerned, there is no policy on spreading misinformation. As a chat app with end-to-end encryption, it is near impossible for the company to track the spread of fake news. Unsurprisingly then, WhatsApp has been one of the main channels used in the circulating of rumours surrounding coronavirus. On the bright side, it was also used to circulate the good news that the British Ministry of Defence is taking over Wembley Stadium to cook the world’s biggest lasagne (Waterson, 2020).
Until recently, Twitter’s policies on the spread of misinformation applied only to large, coordinated efforts, usually backed by state actors (Hern, 2020a). While Facebook encourages users to report posts guilty of spreading fake news, Twitter typically takes no action unless another site rule is broken. This is clear when we report a post on the two different platforms.
Twitter, once described as the “free speech wing of the free speech party”, has been criticised for it’s approach to mitigating the spread of fake news. In light of the coronavirus misinformation scandal, the social network says it will be applying a new broader definition of harm to address content that “goes directly against guidance from authoritative sources of global and local public health information” (Hern, 2020b).
On the topic of free speech, it was recently confirmed by the US Court of Appeals for the Second Circuit that Donald Trump’s habit of blocking critics on Twitter is a violation of the First Amendment (Brown, 2020). Tying things up nicely, the man himself, infamous for his use of the ‘FAKE NEWS’ buzzword, tweeted at the time of writing.
From the perspective of information systems, it seems there is quite a long way to go in the battle against the spread of fake news over social media. Given the continuous developments in the fields of artificial intelligence and machine learning, it’s hard to imagine an automated system for recognising and removing false information won’t be implemented in future.
Instagram have already introduced an A.I. system to detect offensive comments. If the A.I. flags a caption as offensive, it will automatically send a notification to the poster, asking them to reconsider. The company hopes that this pause for reflection will reduce the number of bullying comments posted on the platform (Leighton, 2020). If bullying through social media can be stopped by technology, it’s only a matter of time before fake news can too!
Brown, S. (2020) “Appeals court upholds ruling that Trump can’t block critics on Twitter”, cnet.com, available at:
Hern, A. (2020a) “Fake coronavirus tweets spread as other sites take harder stance”, The Guardian, available at:
Hern, A. (2020b) “Twitter to remove harmful fake news about coronavirus”, The Guardian, available at:
Leighton, H. (2019) “Instagram AI Tool To Flag Offensive, Bullying Captions. But What About Facebook?”, Forbes, available at:
Lyons, T. (2018) “Hard Questions: How Is Facebook’s Fact-Checking Program Working?” available at:
Waterson, J. (2020) “‘They’re building a massive lasagne’: man behind WhatsApp virus spoof revealed”, The Guardian, available at: