Since the 2016 election, partisans have applied the “Fake News” insult to stories that they dislike. Unfortunately, there is such a thing as genuine fake news. Twitter reports there has been fake news in its traffic from thousands of internet propaganda sources. Belatedly, Twitter will notify “677,775 users that they had interacted with fake accounts created by the Internet Research Agency (IRA), a propaganda organization with known links to the Russian government.” The US problem with fake news started during the 2016 presidential race, when Russian operatives ran a covert digital propaganda campaign to sully the reputations of the candidates.
Facebook experienced some of that fake news mixed into its news feed. Fake news flowing over Facebook misled some US prospective voters and some fake news was dangerous enough to trigger violence among rival factions abroad. The Senate has ordered major social networks to fix the problem or face regulations that will fix it. Regulation crafted in anger is very unpleasant to live with.
Reports of “fake” information undermines users’ confidence in what they read and see on the social network sites. That decline in confidence is enough to prompt Twitter and Facebook to repair the flaws that allowed misinformation to contaminate their sites.
In Europe, there are laws that call for prompt take-down of hate speech, stringent consumer privacy, and avoidance of spreading misinformation. Facebook is acutely aware of its occasional failures on those topics. To help squelch hate speech, Facebook is promoting “counter-speech,” a practice of posting positive, affirming ideas that should undermine damage done by hate speech. Responding to British lawmakers, Facebook is conducting another thorough search for evidence that Russia influenced the Brexit vote via propaganda over Facebook. Russia denies meddling in Brexit.
The damage to Facebook’s reputation at home and abroad has pushed it to change its policies. Facebook intends to restore trust in the news that flows across its site. A new process will establish which news sources are the most “trustworthy” and then allow the most trusted sites to deliver the most news in Facebook’s news feed. Aligning a quantity of news to an index of trustworthiness is easy, but establishing a trustworthiness score for a news site is tricky and has overtones of censorship, especially since news sources and Facebook users remain unaware of the trustworthiness scores.
To lessen the appearance of censorship, Facebook will survey a large number of its users to gather their assessment of news sites’ trustworthiness. Mechanically, this is simple, but there have been many criticisms of likely biases in the resulting scores. We are not confident that some users have enough savvy to separate fake from real, especially since some users believed the election-time fake news.
In its favor, Facebook’s process is likely to exclude low-quality “clickbait,” conspiracy theories and bogus stories. Unfortunately, it will not exclude user-posted political drivel, sourced from daily “talking point memos.” It is also unclear if a single set of trustworthiness scores would be universally welcomed in all US states and EU countries.
Facebook is taking additional steps to address the Senate warning and its sagging reputation with users.
Through a Facebook policy called Advert Transparency, every advert posted to Facebook must also be also posted to the advertiser’s webpage. Political adverts will be archived and for federal election adverts, the advertiser will be required to verify its identity and indicate who paid for the advert. Twitter is adopting a similar Ad Transparency strategy for political adverts.
In Europe, Facebook is scattering funds to open community skills hubs in Spain, Poland and Italy. It is setting a digital skills goal for media literacy, and online training for 1 million people and business owners in Europe by 2020. Facebook said it would spend $12 million via its French AI research center to increase its scholarship program, and it will quadruple the number of PhD fellows in its AI Research Paris program.
Consumers will benefit from news that is more trustworthy, the availability of archived adverts, fewer clickbait ads, fewer bogus commercial ads, and the tone improvement coming from counter-speech. It is important for social media to take on this cleanup of misleading content. There are many other challenges headed their way (e.g. entry into streaming media, and social media’s alleged addictive appeal to youth). It would greatly complicate our reaction to those issues if simultaneously, social media were believed to be gateways for propaganda.