Travelers’ Facebook Posts Don’t Belong In The Hands of Border Patrol
By Ashley Houghton, Tactical Campaigns Manager, Amnesty International USA
Imagine every Facebook post you’ve ever written, printed out on sheets of paper. Every tweet, every Tumblr post, every Reddit post, all listed on one page after another, page after page of your history. A post critical of a president five years ago, a rant about a boss you couldn’t stand, a photo of you visiting a local house of worship, all put into the hands of border patrol. Imagine those social media posts — including the most innocuous ones — determining whether or not you can enter the United States.
Yet a proposed new “Extreme Vetting Initiative” by the Immigration and Customs Enforcement Agency (ICE) aims to do just that, and more. In July, ICE asked private companies to help them monitor the internet to assess EVERY individual travelling to the US, and perhaps even green card holders. The objective will be to determine whether to grant a visa — or deport someone — based on whether the individual will be a “contributing member of society” or likely to commit a crime or terrorism — criteria taken directly from Trump’s “Muslim Ban” Executive Order.
To make matters worse, ICE wants this decision to be made automatically, using machine learning technology. In other words, you could be barred from entering the country not because of what you said or did, but because of what an algorithm believes your online activity indicates about your likely future behavior. Today, more than 30 technologists, data scientists and machine learning experts have expressed their grave concerns, saying that in all likelihood, the proposed system would be “inaccurate and biased,” and asking the government to drop the plans. Their call has been echoed by a coalition of more than 100 civil society organizations, including Amnesty International.
This is a deeply dangerous, discriminatory, and nonsensical policy that must be shut down, and fast. Refugees in particular often seek to escape governments that persecute them for their political or social opinions. “Extreme vetting” means that the government can decide whether travelers require additional vetting based on their level of political involvement, whether or not they belong to a faith group, or if they appear critical of U.S. policies. These are all freedoms that the United States aspires to stand by: the freedom of every person to peacefully express themselves without fear of government retribution.
Amnesty and our partners pointed out at the start of the year that failures by big data and data analytics companies to respect their human rights responsibilities could enable the Trump administration to essentially buy a “Muslim registry.” We called on companies to pledge not to do this, and almost none were willing to do so. Now Trump’s “Muslim ban” has morphed into “extreme vetting” and he wants tech companies to build the machines to carry it out. They must not.
Customs officials can already demand your social media handles at the border. With the new ICE proposal, they won’t need to ask — the authorities will already have scanned every corner of your online history, while a machine decides whether you will contribute to the “national interest”.
At this time of social and political divisiveness, “extreme vetting” divides the world a bit further and makes all of us less safe in the process. When arbitrary or flawed data decides who can be allowed into the United States and who is left out, we all lose.
Today, we encourage you to tweet your frustration with the hashtag #DigitalMuslimBan. Share why your data is YOUR data, and why “extreme vetting” hurts us all.
