Facial recognition technology is back in Congress’ crosshairs after IBM Corp., Microsoft Corp. and Amazon.com Inc. each called on lawmakers to establish rules for use of the software and paused sales to police departments in response to the nationwide protests against systemic racism and police brutality.
And although fears abound about what role big technology companies will play in influencing potential new federal standards, surveillance reform activists are cautiously optimistic that meaningful regulation could emerge this year given the growing public awareness of how police and law enforcement employ facial recognition tools in their daily work.
The protesters “are clearly moving tectonic plates on issues of policing and surveillance and law enforcement accountability, generally, and I think we’re seeing a reflection of that in how conversations around facial recognition, specifically, are shifting,” said Evan Greer, deputy director of the nonprofit digital rights advocacy group Fight for the Future.
“We’re in a different moment now than we were the last time this was the focus,” she said.
Over the past few weeks, with momentum from the nationwide protests against systemic racism in response to the killing of George Floyd during an interaction with Minneapolis police, activists have been looking to lawmakers to reform all aspects of policing, including use of surveillance tools like facial recognition.
As a result, lawmakers have already started taking action this month. Democratic Reps. Anna Eshoo (Calif.) and Bobby Rush (Ill.) organized a letter from 35 lawmakers demanding the Federal Bureau of Investigation, the National Guard Bureau, the Drug Enforcement Administration and Customs and Border Protection stop all surveillance of those who participate in peaceful protests. Democrats in the House Oversight Committee, which hosted a series of hearings on facial recognition last year, sent a letter to the Department of Homeland Security on June 5 asking for more information about CBP’s role in surveilling the protests after an agency Predator drone was spotted flying over the Minneapolis protests. And Democratic Rep. Pramila Jayapal (Wash.) started drafting legislation that would place restrictions on the use of the technology for both government and commercial purposes, according to a Politico report.
Meanwhile, a spokeswoman for the House Oversight Committee said the panel is still working on legislation to regulate facial recognition technology.
These quick actions, paired with bipartisan support for regulating this space, has made activists at Fight for the Future, the Electronic Privacy Information Center and the Center for Democracy and Technology optimistic they’ll see meaningful surveillance tech reforms this year — or, at least, a temporary moratorium on the technology’s use while Congress learns more about the data protection, ethics, legality and social implications of the software.
“Facial recognition is one of the very few areas where we’ve seen bipartisan agreement all session long,” said Caitriona Fitzgerald, EPIC’s interim associate director and policy director. “This is an issue that’s kind of been simmering all session, and it’s hopeful that this means Congress will act. Sometimes it does take something like this.”
Although the renewed congressional discussions appear to be in the early stages, activists say they’re watching closely to see what comes from leading lawmakers on the issues — including those in the House Oversight panel, the Judiciary committees in both chambers, as well as privacy hawk Sen. Ed Markey (D-Mass).
“There are a lot of stakeholders who are very interested in this issue, so we’re waiting to see who’s willing to take on the challenge of trying to move legislation on the space,” said Mana Azarmi, policy counsel at CDT’s Freedom, Security and Technology Project.
Facial recognition technology, specifically, has come under congressional scrutiny in recent years because of its encroachment on the privacy of Americans and the racial biases that are implicit in their algorithms. In December, a study from the National Institute of Standards and Technology confirmed that facial recognition technology tends to be better at identifying white people and men after testing 189 software algorithms from 99 developers based on federal government data sets of about 18 million photos of more than 8 million people. Markey has targeted the controversial facial recognition startup Clearview AI, which uses a database based on images scraped from social media accounts and other internet queries.
Until now, congressional efforts have only resulted in a patchwork of bills and a series of educational hearings on the matter. And activists hope to push for something stronger, with Fight for the Future’s Greer saying the group is advocating to either defund federal agency and law enforcement use of the software or ban the technology entirely rather than see any reform that enables the use of facial recognition by creating “rules for the road” for it.
However, the high-profile nature of the issue could encourage the large industry players — which have publicly paused their relationships with police departments and encouraged lawmakers to take up this issue — to take a heavy-handed role in writing these regulations themselves and prompt lawmakers to rush toward the easiest solution, Fitzgerald said.
“What would be a worse outcome is regulations are written hastily and are not sufficient, and then the technology is just deployed with these weak regulations,” Fitzgerald said.
Greer pointed to the lobbying power of Amazon, Microsoft and IBM, as well as the power of intelligence and law enforcement agencies, as major hurdles for activist groups, and she’s concerned that Congress will be “fooled by industry lobbyists.”
“Those are formidable foes who have power and resources,” she said. “We’re not naive about the fact that this will be an uphill battle.”