Tech Experts & Civil Rights Groups to DHS: Automated "Extreme Vetting" Would Be Threat to Constitutional Rights

Today, in two separate letters, a coalition of over 50 civil society organizations - advocates for civil rights, civil liberties, government accountability, human rights, immigrants' rights, and privacy - and a group of more than 50 experts in computer science, engineering, math, and automated decision-making urged the U.S. Department of Homeland Security (DHS) to immediately halt a new "Extreme Vetting Initiative" on the grounds that it will be inaccurate, biased, and a threat to constitutional and human rights.

The Center for Media Justice, alongside other grassroots advocacy organizations, called on companies that attended ICE industry presentations to publicly reject the contract to build the system. In an online petition, they specifically pressed IBM to explain its position, citing the company's pro-immigrant statements in recent policy debates.

The Extreme Vetting Initiative is Immigration and Custom Enforcement's (ICE) plan to monitor much of the internet, including social media, to automatically flag people for deportation or visa denial based on broad, ambiguous criteria. As described in ICE documents, the program will make "determinations via automation" about whether an individual will become a "positively contributing member of society" or "contribute to the national interest" - even though those terms are undefined in American immigration law.

The letter from technology experts, including former senior federal officials in technology policy and luminaries in machine learning and artificial intelligence, explains that "no computational methods can provide reliable or objective assessments of the traits that ICE seeks to measure." It continues: "Algorithms designed to predict these undefined qualities could be used to arbitrarily flag groups of immigrants under a veneer of objectivity," and could "generate a very large number of false positives."

The letter from civil society organizations calls the initiative "tailor-made for discrimination" and argues that it will "undoubtedly chill free expression" by driving travelers and their American friends and family to censor themselves online. They warn that ICE's high-tech proposal "risks hiding politicized decisions behind a veneer of objectivity - at great cost to freedom of speech, civil liberties, and human rights."

The petition to IBM says that "IBM has a responsibility, beyond lip-service, to stand up to Trump and his administration's plans to discriminate against immigrants. That means not being complicit with agencies like ICE and not helping them build racist tools that only further marginalize immigrants."

The organizations forming part of this effort will hold a teleconference at Noon ET / 9am PT for members of the press. To participate, call 888-632-3385.

Speakers include: Alvaro Bedoya, Executive Director of the Center on Privacy & Technology at Georgetown Law; Rachel Levinson-Waldman, Senior Counsel to the Brennan Center's Liberty and National Security Program; Joshua Kroll, Computer Scientist and Postdoctoral Scholar at the School of Information at the University of California at Berkeley; Brian Root, Quantitative Analyst at Human Rights Watch; and Steven Renderos, Organizing Director at the Center for Media Justice.

Click here to read the full list of signatories on the civil society letter.

To review the ICE contract documents for the Extreme Vetting Initiative, and for other background, visit the Brennan Center and the Center on Privacy & Technology's resource page.