Lydia Laurenson Polarisation and Peacebuilding Strategy on Digital Media Platforms: Current Strategies 7 nonstop at horrifying images) and gaps in the moderation process (gaps that are hard to report to Facebook because the moderators are not “real” Facebook employees). But excitingly, in mid-2019, Kate Klonick reported in Slate that Facebook has taken these concerns to heart and is setting a new bar for how content moderators are treated. Removal from Facebook (“deplatforming”) Facebook has been removing fake accounts from Facebook for quite a while and has enforced its “real name policy” as well (i.e., even if an account is not fake, it can still get taken down if it is not attached to the user’s “real name”). Recently, it has come under more pressure — and has shown more willingness — to remove or “deplatform” real accounts. For example, in early 2019, The Guardian published a piece by Julia Carrie Wong on how Facebook has chosen to “ban four ethnic armed Myanmar-based groups from its site,” all of which oppose the Myanmar government that is currently in place. (Indeed, although the industry historically was extremely reluctant to censor anything, Silicon Valley has been tilting more towards censorship than it once did — sometimes in problematic ways, as the Electronic Frontier Foundation points out.) While some observers praise Facebook for banning distasteful figures like American white supremacists, it is not clear how (or whether) Facebook will create a consistent, transparent, and just policy for who gets deplatformed. Removal of creator income (“demonetisation”) In 2018, Facebook began rolling out tools to enable users to send money to their favourite creators. (For example, a user could send money to an artist who posts videos on Facebook.) While rolling out those tools, Facebook also announced a policy about which content is eligible for monetisation. This means that a creator who earns income from Facebook can have that income taken away if they violate Facebook’s monetisation rules: they can be “demonetised.” Facebook demonetisation seems uncontroversial so far, perhaps because Youtube has already been monetising and demonetising content for years. Advertisers have also insisted on not showing their ads next to controversial content since time immemorial. Both these factors mean that Facebook did not have to develop brand-new policy or procedure for this. (On the other hand, it took several years for Youtube’s demonetisation policies to occasion blowback, as described by Peter Kafka in Recode back in 2016, so there may be blowback yet to come.) Political labeling In 2017 and 2018, Facebook created tools called Town Hall and Candidate Info. These are clearly delineated spaces on the platform where candidates are labeled as such, and users exposed to a candidate message have the option to look at the opposing candidate’s message. (The labeling is reminiscent of Reddit’s “flair” concept, though it appears to be less driven by human moderators.) These spaces are intended to make political maneuvering more transparet and have attracted little criticism so far — although it will be hard for Facebook to decide who gets designated as “legitimate” politicians in less stable countries like Myanmar or Cambodia.Fact-checking partnerships Facebook works with outside partners to fact-check news on the platform, and this effort

Select target paragraph3