A look inside the alleged Facebook internal training and policy manuals leaked to The Guardian newspaper late last week reveals an often-conflicting set of rules Job himself might not have the patience to untangle.
According to the documents obtained by The Guardian, Facebook’s moderator instructions include:
Remarks such as “Someone shoot Trump” should be deleted, because as a head of state he is in a protected category. But it can be permissible to say: “To snap a bitch’s neck, make sure to apply all your pressure to the middle of her throat”, or “fuck off and die” because they are not regarded as credible threats.
Videos of violent deaths, while marked as disturbing, do not always have to be deleted because they can help create awareness of issues such as mental illness.
Some photos of non-sexual physical abuse and bullying of children do not have to be deleted or “actioned” unless there is a sadistic or celebratory element.
Photos of animal abuse can be shared, with only extremely upsetting imagery to be marked as “disturbing”.
Videos of abortions are allowed, as long as there is no nudity.
Facebook will allow people to livestream attempts to self-harm because it “doesn’t want to censor or punish people in distress”.
Anyone with more than 100,000 followers on a social media platform is designated as a public figure – which denies them the full protections given to private individuals.
Facebook rules regarding sexuality are perhaps the most arcane. For example, the manuals indicate “moderate” sexual activity — “open-mouthed kissing, clothed simulated sex and pixelated sexual activity” — between consenting adults are allowed. Nudity is prohibited in contemporary “digital” media, but is fine in paintings, sculptures, and other forms of what Facebook calls “handmade” art. “Handmade” art may even be sexually explicit, to a point.
That’s a fine line, and Facebook mods must evaluate each instance on a case-by-case basis. Last year, the social network came under fire for removing the Pulitzer Prize-winning news photograph of a naked little girl running from a Vietnamese village bombed with napalm. The company also has been criticized for removing images of women breastfeeding.
Sexual language is allowed in some instances but not in others. Examples of approved explicit language include “I’m gonna eat that pussy,” “Hello ladies, wanna suck my cock?” and “I’m gonna fuck you.” Facebook moderators are warned to draw the line before the language goes into detail about whatever actions the speaker has in mind.
Images of child sexual abuse are prohibited (by law), but images of other kinds of child abuse are OK on Facebook under the reasoning someone may be able to identify and help the children depicted.
“We do not action photos of child abuse,” Facebook notes in its guidelines. “We mark as disturbing videos of child abuse. We remove imagery of child abuse if shared with sadism and celebration.”
The same sort of rule applies to revenge porn, which is deemed revenge porn only if the poster obviously intends to shame or harm the individual depicted and posts the images “with sadism and celebration.”
With more than 2 billion users accessing the social network daily, the moderation staff of 4,500 humans and a software program or two are outmanned. One moderator told the Guardian they often have only 10 seconds to make a decision about any given post. Earlier this month, after users live-streamed a murder, a couple of suicides and an honor killing, Facebook announced plans to add another 3,000 moderators.
Still, considering current moderators investigated 54,000 reports of revenge porn, sexual extortion and child sexual abuse in January alone (and removed more than 14,000 of those) and they review 6.5 million reports of fake accounts a week, the social media giant may be waging a quixotic war to banish evil from a platform designed to celebrate free expression.
