US mid-terms: judgement day for social networks

08 Nov 2018
   —  by Giles Brown

This week’s US mid-term elections served as a litmus test of President Trump’s popularity – a vindication or condemnation of his time so far in office.

With such huge interest and engagement, the world’s social media giants were under tremendous pressure to clean up their act after the meddling of the Russian troll factories in the 2016 Presidential elections.

Social360 wrote a few months ago of the War Room set up by Facebook to dig out and eradicate false news and fake accounts attempting to influence voters.

In the run up to the mid-terms, Reddit took similar action. The community-based message board is the sixth most popular social media platform globally, more popular than Twitter. Its users are reportedly “young and male”, with over 54% coming from the US. Reddit is the hunting ground of the Trump supporter.

The large message groups that exist on Reddit have the power to spread fake content with ease, especially given the long-form nature of opinion expressed through the site.

The platform identified almost 1000 false accounts which had made around 14,000 posts. Its CEO Steve Huffman wrote “I believe the biggest risk we face as Americans is our own ability to discern reality from nonsense, and this is a burden we all bear.”

Twitter also deleted around 10,000 automated accounts on advice from the Democratic Congressional Campaign Committee which was launched in the wake of the vast amounts of false information circulated about Hilary Clinton during the 2016 campaign.

And just this week Facebook has disclosed that it removed over 100 accounts (including some on its sister account Instagram) on the eve of the mid-terms on evidence from the FBI. The following day, the Internet Research Agency – currently under investigation by Robert Mueller for widespread interference in the 2016 election – claimed responsibility.

Regardless of the election success that both the Democrats and President Trump are now claiming, can the social media behemoths also feel assured they succeeded in eradicating external voter influencing campaigns? Were the mid-terms a victory for Facebook, Reddit and Twitter?

The BBC is reporting that the defences of the social networks have been held “for now”, noting that the measures that have been put in place seemed to have stopped efforts to influence voters “at source”. They pick up on sentiment that sources may instead try to spread misinformation after the vote in an attempt to discredit the result.

With the next Presidential election now two years away, how can the social media networks ensure that their checks and balances are more watertight than ever? It begs the question whether a machine can ever prove a source’s authenticity with absolute conviction.

When Social360 was founded ten years ago, we recognised the noise and false information that was clouding the social media landscape. We built our model on the recognition that algorithm-based social listening can combine with human insight to deliver analysis that can be trusted as reliable and authentic.

With no foreseeable end to the new methods by which automated feeds are polluting genuine content, our model is more relevant than ever.

To find out more about how our model of robust online monitoring can help you manage social media conversation, do read our paper published earlier this year and get in touch.

Giles Brown