Are killer robots now legal in San Francisco?

It might seem like the starting point of a dystopian sci-fi movie plot, but police in San Francisco have been granted permission to use lethal robots to kill people in emergency situations.

City politicians voted earlier this week to allow the San Francisco Police Department (SFPD) to deploy robots in “extraordinary circumstances” and in a “virtually unimaginable emergency”, said Aaron Peskin, dean of the city’s Board of Supervisors.

The board voted 8-3 in favour of allowing police to use the remote-controlled robots after “an emotionally charged two-hour debate”, said Al Jazeera. The majority vote was despite “strong objections from civil liberties and other police oversight groups”. Critics argued it could “allow police to kill people too easily”, said the San Francisco Chronicle.

Subscribe to The Week

Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.

SUBSCRIBE & SAVE

Sign up for The Week's Free Newsletters

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

Sign up

A decision by the board was required by a 2021 law that specified that “law enforcement agencies get approval from their governing bodies” for the use of certain weapons. The board has the power to “accept or reject how the equipment is used”, wrote the Chronicle.

The SFPD said it would use the robots to kill suspects as “an extra tool to save the lives of officers and save the lives of people”. SFPD assistant chief David Lazar cited October 2017’s mass shooting at a Las Vegas music festival, in which 60 people were killed and hundreds injured, as an example of a scenario in which a robot could be used to neutralise a suspect.

What kind of robots are they using?

The SFPD currently has “12 functioning robots” that are all remote-controlled (rather than autonomous) and primarily used to “gain situational awareness and survey specific areas officers may not be able to reach”, according to NPR. Other key functions include helping “investigate and defuse potential bombs, or aid in hostage negotiations”.

There were “no plans to outfit robots with a gun”, the SFPD told the Chronicle, but some of the robots could be “equipped with explosive charges to breach structures containing violent suspects”. Some of the department’s robots are a model called Remotec F5A, the paper said, which can “climb stairs, lift over 85 pounds, overcome curbs, probe any hazmat situations, and self-right when they’re flipped over”.

Who else is using robots?

While remote-controlled bomb disposal robots are widely used, robot dogs have had a “resurgence in recent years, particularly among law enforcement agencies”, wrote Gizmodo. These dog-shaped robots, developed by Boston Dynamics, are “ideal surveillance tools for law enforcement in search of roaming sentries” and are used to inspect dangerous areas before people enter.

“Police forces spanning at least three continents” are using dog robots, according to Gizmodo, including previously the New York Police Department (NYPD), Dutch police and authorities in Singapore.

The NYPD was forced to “cut its $94,000 contract with Boston Dynamics short” after its highly publicised use of the robot dog was “poorly received”. Shortly after, the New York City Council introduced a bill that “would prohibit police from arming robots used in their law enforcement operations”, said Human Rights Watch.

Have robots been used to kill suspects?

Police in Dallas in 2016 used a robot to kill a suspect for the “first time in history”, said The Guardian. On that occasion, police equipped a bomb disposal robot “with an explosive device on its manipulator arm to kill a suspect”. The suspect was a sniper who had already killed five police officers and injured seven others.

The adaptation of the robot to become lethal was the “kind of repurposing [that] has until now been limited to the military”.

What are the problems?

In San Francisco, some of those on the board “remained deeply skeptical that police need the extraordinary ability to kill someone with a robot”, wrote the San Francisco Chronicle.

One of the city’s officials, Dean Preston, argued that there was “serious potential for misuse and abuse of this military-grade technology, and zero showing of necessity”.

More widely, “the idea of robots being legally allowed to kill has garnered some controversy”, said NPR, pointing to an open letter signed by leading robotics companies that said that the “general purpose robots should not be weaponized”.

Ryan Calo, a professor at the University of Washington who studies robotics, added that if a robot used by police accidentally kills a suspect it can be extremely problematic, telling NPR it would be “difficult to disentangle who is responsible”. He added that while robots can be attractive to use for police forces, because it limits “putting themselves in harm’s way”, it also “feels so deeply dehumanising and militaristic”.