With fake news dominating the real news, telling what's genuine on social media is a big concern. It's all about social bots. But what are they and how are they affecting our political systems?
With social media playing an ever-growing role in the way we all perceive the world, the manner in which websites like Facebook, Twitter and others portray information to consumers has never been more important.
Where everyday citizens used to judge the importance, popularity or credibility of politicians, economic models or ideological movements based on professional news reports, polling or simply through word of mouth, they are now swayed by what may be trending or deemed accepted on their social media feeds.
This is where social bots come in. Using pre-programmed bots from a variety of companies, from BMW to Spotify, you can sign up to have a bot save every song you've recommended on a music app, automatically post your Instagram photos to a Twitter account or even open your garage door when your car pulls into the drive way. Yet some make their own bots with malicious software.
The most obvious manifestation of these projects are on Facebook and Twitter, where fake accounts are setup with thousands of fake followers to create the illusion of authenticity. Once partnered with the software of a certain type of bot, these fake accounts can target people with tweets or posts that contain specific messages or indeed retweet or favorite a person's tweets to boost their own popularity.
Manipulation of the facts is a hot topic in today's political environment. German Justice Minister Heiko Maas said late last year that he fears false information or "fake news” may spread misinformation ahead of the Bundestag elections this autumn.
And he has good grounds for worrying. Although fake news is written and published by humans, it is often distributed and supported by millions of fake accounts on popular sites like Twitter, Facebook, Reddit, and through email. A study by Buzzfeed News shortly after Donald Trump's election victory showed that fake election stories gathered more engagement on the social media platform than genuine stories from 19 major news outlets combined.
A report conducted by "Political Bots", a team of researchers from leading universities around the world, found that as many as 30 million accounts on Twitter belonged to fake users. It also found that ahead of the US Presidential elections pro-Trump bots, which promoted fake news stories, outnumbered pro-Clinton ones by a factor of 5:1.
"Bots - algorithmically driven computer programs designed to do specific tasks online - have invaded political conversations worldwide," warns the report. "The pervasive use of such human-software hybrids, and the obscure and often discriminatory nature of the algorithms behind them, threaten to undermine the political potential - organizational, communicative, and otherwise - of social media systems."
Bots are also used on the internet to perform DDoS attacks, which can bring down entire systems
Combating the issue
Although companies like Hoaxmap have been fighting fake news since February 2016, social media companies, authorities and governments are now trying to catch up.
Facebook recently announced an initiative to tackle fake news on its platform, with new tools to make it easier for users to flag a fake news story on their timeline. Partnering with Berlin company Correctiv, a nonpartisan, nonprofit investigative journalism collective, Facebook hopes to use third-party face checking to review flagged material.
The German government wants to tackle the problem head-on, with a new body linked to Interior Ministry, called the "Center of Defense Against Disinformation."
Researchers at Cambridge University, meanwhile, say a "psychological vaccine" inoculate people against fake news - but, as with other vaccines, that could be decades in the making.