A vote by San Francisco city supervisors in support of allowing police to use robots to kill people in emergency circumstances has drawn criticism from experts in law enforcement and security who say it could lead to dangerous and unintended consequences.
The San Francisco Board of Supervisors voted Tuesday to advance an ordinance that includes a controversial policy that would give city police the ability to deploy robots to use lethal force in extraordinary circumstances “when risk of loss of life to members of the public or officers is imminent and officers cannot subdue the threat after using alternative force options or de-escalation tactics.”
After a contentious debate, the members of the board voted 8 to 3 in support of the policy, which will need to pass another vote by the board on Dec. 6 and then be signed by the mayor for it to become law.
The vote came about after California passed a law last year requiring law enforcement departments to seek approval for use of military-style equipment.
“I don’t think that we’ve done enough research” or that “we’ve really been thoughtful about this,” said Kirk Burkhalter, a law professor at the New York Law School and a retired detective with the New York City Police Department.
Burkhalter said he believes it would be extremely rare for a police response to necessitate the use of a robot capable of killing. But in taking the step toward approving such, Burkhalter said he believes police departments around the country will follow San Francisco’s lead, making it commonplace to deploy robots to use deadly force.
“Once we go down this road, it’s very difficult to pull back,” he said.
Matthew Guariglia, a policy analyst with the Electronic Frontier Foundation, a nonprofit organization that advocates for civil liberties in the digital realm, criticized the measure for using what he said is vague language that makes it “not at all unreasonable for us to assume that in standoffs with people in the midst of a mental health crisis or something police will now feel empowered to send out robots that are equipped with explosive charges.”
Paul Scharre, the vice president and director of studies at the Center for a New American Security, a think tank focused on national security issues, said the San Francisco proposal was “the exact opposite of what we should be using robots for in a policing context.”
Scharre, who is also the author of “Army of None: Autonomous Weapons and the Future of War,” said the advantage of using a robot is that it creates a safe distance between police officers and armed suspects, and thus police should try to “find ways to use that increased distance as a way to build more options for law enforcement to use” instead of lethal force.
The vote in San Francisco comes more than six years after police in Dallas used a robot with an explosive device to kill a suspect who was holed up inside a parking garage after an ambush in which five officers were killed and seven others were injured, along with two civilians.
“We are pretty far away from this being an effective, thoughtful tool to be used in anything but extreme circumstances like Dallas,” Burkhalter said, adding that he feared the San Francisco vote was getting “ahead of ourselves.”
The San Francisco Police Department said in a statement that the policy allowed for the ability to use “potentially lethal force in extreme circumstances to save or prevent further loss of life.”
“The use of robots in potentially deadly force situations is a last resort option. We live in a time when unthinkable mass violence is becoming more commonplace,” San Francisco Police Chief William Scott said in the statement. “We need the option to be able to save lives in the event we have that type of tragedy in our city.”
The police department acquired robots between 2010 and 2017 and “does not own or operate robots outfitted with lethal force options and the Department has no plans to outfit robots with any type of firearm,” according to the statement.
The statement said the robots are used primarily in bomb situations or other incidents in which police officers may need to keep a safe distance. But in “extreme circumstances” those robots could be used to deliver an explosive to breach structures where there is an armed suspect, and could incapacitate a suspect or cause injury or be lethal.
The department said only the chief of police, assistant chief of operations, or deputy chief of special operations could authorize the use of robots to potentially kill a suspect.
During the board of supervisors meeting this week, David Lazar, the assistant chief of the San Francisco Police Department, gave the 2017 Las Vegas shooting as an example of when police could have benefited from the use of robots with the ability to kill. In that case, a gunman fired a barrage of bullets from the 32nd floor of a hotel, killing 60 people in one of the deadliest mass shootings in the U.S.
Dean Preston, a supervisor on the board who voted against the measure, said he has received a “really widespread outpouring of opposition” to the proposed policy from constituents across the city.
Preston said that the country has been in a national conversation about police violence and racial disparities in policing since the death of George Floyd in 2020 and that the vote “is certainly an indication that we’ve really strayed pretty far from that conversation we need to be having.”
Preston cited the San Francisco Police Department’s own data that found that in 2021 police used force on Black people 12 times more than white people.
“We run a very serious risk of misuse by police of a robot to inflict deadly force,” he said. “In the United States, we have a number of examples of abuse of that power even without robots.”
Preston referred to a 1985 bombing in which Philadelphia police dropped a bomb onto the compound of the Black organization MOVE, killing six members and five of their children, and destroying 65 homes in the neighborhood.
He also highlighted an incident from June 2021 in which Los Angeles police detonated a massive cache of illegal fireworks, causing a “catastrophic” blast that injured 17 people and damaged dozens of properties.
Preston said he hoped that outrage following the first vote in San Francisco would sway more of his fellow board members to vote against the measure Tuesday.
Matt Dorsey, a member of the board who supported the policy, said during the meeting this week that he was not comfortable putting an explicit restriction on robots using deadly force that was “unique or at least rare among cities in the United States.”
“I want to make sure that we have state of the art equipment, and God forbid if we are ever in a situation where we have a terrorist attack or a mass shooting event, that we are not depriving ourselves of access to something that will protect public safety,” he said.
Catherine Stefani, a member of the board, added that police needed to respond to “the sad situation of the gun violence epidemic in our country.”
“These military-style weapons are on our streets,” she said. “So when people talk about the militarization of our police force without context as to what’s happening around us, it’s really unfair.”
But Scharre said he found the robot proposal at odds with the fact that in San Francisco, police are not authorized to carry stun guns.
“If you don’t trust someone with a Taser, would you let them have a robot armed with explosives?” he said.