When social media platforms enforce their content moderation rules unfairly, it affects everyone’s ability to speak out online. Unfair and inconsistent online censorship magnifies existing power imbalances, giving people who already have the least power in society fewer places where they are allowed a voice online.
President Donald Trump and some members of Congress have complained on Twitter that the people most silenced online are those who share the President’s political views. So the White House launched this week a website inviting people to report examples of being banned online because of political bias.
The website leads users through a series of forms seeking their names, citizenship status, email address, social media handles, and examples of the censorship they encountered. People are required to accept a user agreement that permits the U.S. government to use the information in unspecified ways and edit the submissions, raising a natural concern that any political operation will selectively disclose the results.
For this reason, we have sent a FOIA request to the Executive Office of the President, Office of Science and Technology, for all submissions received. And because we are concerned that the White House's website asks for a lot of personal information without sufficient privacy protections, we have asked that first name, last name, email address, phone number and citizenship status be redacted.
It’s troubling to see people being asked whether they are a U.S. citizen or a permanent resident in light of the administration’s hard line immigration policies. It raises legitimate questions about how that and other required personal information may be used and who the government really wants to hear from.
There’s no question that social media platforms are failing at speech moderation—that’s why EFF has been monitoring their practices for years and working with other civil society organizations to establish guidelines to encourage fairness and transparency. Starting in 2014 EFF has been collecting stories of people censored by social media at Onlinecensorship.org. The reports we've received over the years indicate that all kinds of groups and individuals experience censorship on platforms, and that marginalized groups around the world with fewer traditional outlets and resources to speak out are disproportionately affected by flawed speech moderation rules.
While we agree that mapping online censorship is important, we should remember that surveys like these will, at best, give an incomplete picture of social media companies' practices. And let’s be clear: the answer to bad content moderation isn’t to empower the government to enforce moderation practices.
Rather, social media platform owners must commit to real transparency about what speech they are removing, under what rules, and at whose behest. They should adopt moderation frameworks, like the Santa Clara Principles, that are consistent with human rights, with clear take down rules, fair and transparent removal processes, and mechanisms for users to appeal take down decisions.