Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
×
Robotics

San Francisco Police Seek Permission For Its Robots To Use Deadly Force (engadget.com) 143

An anonymous reader quotes a report from Engadget: The San Francisco Police Department is currently petitioning the city's Board of Supervisors for permission to deploy robots to kill suspects that law enforcement deems a sufficient threat that the "risk of loss of life to members of the public or officers is imminent and outweighs any other force option available to SFPD." The draft policy, which was written by the SFPD itself, also seeks to exclude "hundreds of assault rifles from its inventory of military-style weapons and for not include personnel costs in the price of its weapons," according to a report from Mission Local.

As Mission Local notes, this proposal has already seen significant opposition from both within and without the Board. Supervisor Aaron Peskin, initially pushed back against the use of force requirements, inserting "Robots shall not be used as a Use of Force against any person," into the policy language. The SFPD removed that wording in a subsequent draft, which I as a lifelong San Francisco resident did not know was something that they could just do. The three-member Rules Committee, which Peskin chairs, then unanimously approved that draft and advanced it to the full Board of Supervisors for a vote on November 29th. Peskin excused his decision by claiming that "there could be scenarios where deployment of lethal force was the only option."

The police force currently maintains a dozen fully-functional remote-controlled robots, which are typically used for area inspections and bomb disposal. However, as the Dallas PD showed in 2016, they make excellent bomb delivery platforms as well. Bomb disposal units are often equipped with blank shotgun shells used to forcibly disrupt an explosive device's internal workings, though there is nothing stopping police from using live rounds if they needed, as Oakland police recently acknowledged to that city's civilian oversight board.

This discussion has been archived. No new comments can be posted.

San Francisco Police Seek Permission For Its Robots To Use Deadly Force

Comments Filter:
  • What's next? ED-209?
  • Question: (Score:5, Insightful)

    by Dj Stingray ( 178766 ) on Wednesday November 23, 2022 @08:09PM (#63075654)

    Why would a robot need to protect itself using deadly force? It's just a machine.

    • Re:Question: (Score:4, Insightful)

      by Rujiel ( 1632063 ) on Wednesday November 23, 2022 @08:31PM (#63075718)
      It's not protecting itself, it's protecting the wealthy like a good little cop does
    • Not itself, but a hostage situation with traps is a realistic scenario. Sure, you can just go full Putin and use poison gas but that causes even more whining.

      • Re: (Score:2, Funny)

        A "full Putin" would have to include the robot dropping its weapons at some point and running away...

      • That's not a realistic scenario at all except for in a movie.
      • Not itself, but a hostage situation with traps is a realistic scenario.

        There are many other more serious problems the SF police should be worrying about before this one.

      • a hostage situation with traps is a realistic scenario.

        How realistic? Can you give us a list of occurrences? Movies don't count.

    • Re:Question: (Score:5, Insightful)

      by Arethan ( 223197 ) on Wednesday November 23, 2022 @09:03PM (#63075834) Journal

      It isn't to protect the robot.

      From TFA:

      permission to deploy robots to kill suspects that law enforcement deems a sufficient threat that the "risk of loss of life to members of the public or officers is imminent and outweighs any other force option available to SFPD."

      They want the option to send in a drone to snuff out the "bad guys", so they don't need to risk their own lives in the process of ending the threat. The military has been doing this with reaper drones in the middle east since Obama. I'm sure SFPD isn't asking for rocket equipped aircraft, but the premise is the same -- they want a significant tactical advantage.

      The question is: Do you trust them to responsibly wield that power?

      If we look to the military use of drone executions for datapoints on this, we can surmise that this will greatly increase the number of police performed executions. Zero risk to personnel, coupled with a good chance of success, will make this a very attractive looking option in their playbook. And the more they do it, the better they'll get at it, until it becomes the defacto response because it "just works".

      On the other hand, the alternative is that the FBI fills the building with tear gas, and an errant flame causes explosive ignition. We've seen that one play out a few times now.

      Sometimes all that is available are the shitty options.

      • Re:Question: (Score:5, Insightful)

        by ArchieBunker ( 132337 ) on Wednesday November 23, 2022 @09:30PM (#63075900)

        Maybe if the police are too scared to do their jobs then they shouldn’t be police. Roofers and crab fishermen have higher fatality rate than the pussy po po.

        • by Sloppy ( 14984 )

          You might be surprised to learn that roofers and fishers also have incentive for their own jobs to become less risky. I think if a roofer or fisher wanted to introduce a new remote-control tool to make their work safer, it would be as welcomed as most other technological developments.

          While I don't use bluetooth keyboards and mice for safety purposes, if my computers had been more physically dangerous, I think I would have adopted bluetooth even sooner.

          I wouldn't expect policing to be a special case in that

      • by rsilvergun ( 571051 ) on Wednesday November 23, 2022 @09:33PM (#63075910)
        Is the alternative that the FBI fills the building with tear gas? It's a fucking building. You surround it and wait the person out. If they haven't already killed their hostages then they're probably not going to.

        And even better nine times out of 10 when you have a hostage situation it's either because a dude with multiple DV convictions somehow still has a gun and has it pointed at his girlfriend or because the cops botched to no knock warrant.

        I get that we can't do anything about the guy with the DV convictions because if we took the guns away from every guy who smacked his girlfriend around it would have a noticeable effect on gun sales. But there's absolutely no reason for these stupid ass no knock warrants. Every time you see one it's because the cops want to play with their toys. The person they're after could have just as easily been picked up off the street when he was out for a jog like the guy at Waco.

        Why is it that we can't even imagine not living in a violent militaristic police state? That the options are never, don't let cops kill us, but maybe let cops kill fewer of us?
        • by cstacy ( 534252 )

          I get that we can't do anything about the guy with the DV convictions because if we took the guns away from every guy who smacked his girlfriend around it would have a noticeable effect on gun sales.

          What it would have is a noticable effect on retaining cops on the force.

      • On the other hand, the alternative is that the FBI fills the building with tear gas, and an errant flame causes explosive ignition. We've seen that one play out a few times now.

        Sometimes all that is available are the shitty options.

        Or they could get a better tear gas. There are options.

      • Re:Question: (Score:4, Informative)

        by Anonymous Coward on Wednesday November 23, 2022 @11:17PM (#63076078)

        The question is: Do you trust them to responsibly wield that power?

        If the police want to live in a country where the constitution grants them the power to unilaterally execute innocent civilians... Perhaps they should move to one.

        Yes, a "suspect" is otherwise known as an innocent civilian.
        Here only courts have the power to judge a suspect as guilty of a crime and sentence them to capital punishment.

        And no I don't even trust the courts with the power of capital punishment, but at least they are acting completely within the law when they do so.

      • In a way, I'd love to see drones like this become the norm. Maybe "I feared for my life" will no longer be an acceptable excuse for shooting some poor unarmed black kid when the actual police officer is safely back at the precinct and piloting a drone.
      • Allow them to use such robots, if the commander who approves the deployment takes personel liability if there is a cock up during the ops.

        And furthermore, they have to release all footage from that op to the public within 24 hours - maybe after removing sensitive footage.

        If the above two points are agreed, sure, go ahead.

    • It's not about protecting the robot, it's about protecting the police officers.

      See this example:

      The death of a suspect in the Dallas police shootings marks the first time U.S. police officers have used a robot to kill someone, according to Texas and national experts.

      Hourslong negotiations with the man broke down into an exchange of gunfire, Dallas Police Chief David Brown said at a news conference Friday morning. At that point, the officers deployed a robot armed with an explosive.

      "We saw no other option but to use our bomb robot and place a device on its extension for it to detonate where the suspect was," Brown said.

      [...]

      https://www.texastribune.org/2... [texastribune.org]

    • Re:Question: (Score:4, Insightful)

      by ArchieBunker ( 132337 ) on Wednesday November 23, 2022 @09:27PM (#63075894)

      So the cops can have even less liability (if that’s even possible) when they kill an innocent person. Not or fault it was the robot.

    • It is not an autonomous robot. It is a remote controlled vehicle which has a shotgun. The shotgun is usually loaded with a blank (which fires very hot gas, usually sufficient to disrupt small suspicious objects), but which can be loaded with any type of shotgun cartridge, including shot or sabots. These machines have been in use world wide for decades and some sensitive snowflake finally noticed the shotgun.
  • To give them that permission? I mean I get the billionaires will happily do it because they can use death robots to put us down. We already do that with drones in countries that don't do what we want them to. But why is it we refuse to take power away from our "betters". Are we really this fucking stupid?
    • Yes, billionaires are totally out to put you down with death robots, the first part of their plan is to make it so that the SFPD can deploy them first, even though what SFPD is asking for is something that has already been done before and was used to take down an active shooter. But that's not enough, they need the SFPD to have it before they can bring their plan to fruition. Totally not a conspiracy theory. Now that they have a billion dollars, they already have everything they need, so what use are you to

  • by BrendaEM ( 871664 ) on Wednesday November 23, 2022 @08:11PM (#63075664) Homepage
    People are idiots.
  • As long as citizens can also use robots to engage with police too.
  • They come in matched pairs.
  • Imagine (Score:5, Insightful)

    by phantomfive ( 622387 ) on Wednesday November 23, 2022 @08:19PM (#63075688) Journal

    Imagine if the SF police did something about broken car windows.

    • Why should they? The courts will not convict, the DAs will not prosecute, and state law prevents it. Based on the last election in CA, that's exactly how the populace wants it.

  • by RightwingNutjob ( 1302813 ) on Wednesday November 23, 2022 @08:22PM (#63075702)

    It's not a *real* kill if you have to resort to such artiface. Your bare hands and teeth are what God gave you...use them and savor the flesh of your vanquished foe!

  • by kyoko21 ( 198413 ) on Wednesday November 23, 2022 @08:25PM (#63075710)

    Who knew OCP would move its headquarters from Detroit to San Francisco?

  • by mark-t ( 151149 ) <markt AT nerdflat DOT com> on Wednesday November 23, 2022 @08:36PM (#63075736) Journal
    A robot should be more than able to incapacitate a person without killing them.
    • In certain hostage situations, that might be impossible, especially from hundreds of yards away. I can think of many scenarios in which it might be the only option to save many lives.

      • by mark-t ( 151149 )

        From hundreds of yards away, what is the need to use a robot in the first place, exactly? Why can't a person be in control of whether or not deadly force is required?

        I have a big problem with leaving a decision on whether a human being lives or dies up to a machine, because while humans have to face consequences for their choices. Machines don't.

        • by uncqual ( 836337 )

          I don't think there's any suggestion here that the SFPD would let the robot(s) make the decision about if/how the robot should deploy lethal force.

          Usually I'm sure these robots would be under complete control of a human via camera and remote control.

          Even if there was a situation, perhaps due to a shooter being in what is effectively a Faraday cage, where a robot was preprogrammed to take a certain path and then do something like explode a charge, that really wouldn't be much different than a human lobbing g

        • If you would actually RTFA, then you should notice that the device is a remote controlled bomb disposal vehicle, not an autonomous robot.
      • I can think of many scenarios in which it might be the only option to save many lives.

        Do go on.

        • You have a hostile shooter who has displayed competence in shooting, who may/probably has an explosive suicide vest on, holed up in an area where you may not easily(or even with difficulty) dislodge them with gas, lobbed grenades, or long range rifle fire. Where they still pose a risk to the public at large, perhaps due to availability of spots they can use to shoot outside.

  • by swell ( 195815 ) <jabberwock@poetic.com> on Wednesday November 23, 2022 @08:41PM (#63075748)

    On one hand, it makes sense. A well armed crazy should not be confronted by a human if possible. Use a robot as is already done in certain negotiation situations.

    On the other hand you have the liability distinction. When a human cop shoots a civilian, he might be held liable for a bad decision. When a robot shoots a civilian, who is responsible? The AI programming? The programmer? The committee of cops who OK'd the shooting? Most likely they are all relieved of responsibility due to the difficulty of laying blame. Cops love it when they can't be held responsible!

    • From what I've seen, the same people who always end up paying(taxpayer). Sure the officer may lose their job, possibly even go to jail if really egregious, but almost certainly the city will pay a 7-8 figure payout to the family. Always.
    • It's a weapon. So how about holding liable the estate of the perp who brought about the situation? How often do gun manufacturers have to be liable for deaths their weapon causes? The fact now is situations like Uvalde and Parkland where the cops are too afraid to bust in, there needs to be an option like to save lives without risking more lives.

    • by uncqual ( 836337 )

      I suppose in many cases the same logic could be used that will be used with "self driving" cars that injure or kill a human who was not 100% at fault.

    • by GuB-42 ( 2483988 ) on Wednesday November 23, 2022 @10:23PM (#63076006)

      These are remote controlled robots, no AI here. The officer controls the robot, acquires the target and pulls the trigger. The only difference is that he does it from a safe place.

    • Remember, many robots are teleoperated, not fully autonomous. With armed reaper drones, for example, somebody still pulls the trigger.

      With "fire and forget" munitions, which range from as "smart" as a cruise missile down to, technically speaking, even a plain lead bullet fired from a gun, you look towards the operator - the one who made the final launch decision.

      As for the cops, sure, they love it when they can't be held responsible, but I gotta point out that it's already an uphill battle to do that for a

  • (that's what the word "suspect" means, right?) to be shot by a robocop?

    Better go back to the "perp" terminology before implementing this.
  • by gweihir ( 88907 ) on Wednesday November 23, 2022 @08:56PM (#63075802)

    So they want to kill people with no risk to themselves? Killing people must never be easy and must never be risk-free or it becomes far too easy to do.

    A certain amount of personal courage is very much non-optional for a police-person. If you do not have it, then look for another job. It is not like being a cop is particularly dangerous job in the first place.

    • Re: (Score:3, Insightful)

      by uncqual ( 836337 )

      If a person is clearly posing a serious threat to the life of others, why should it be necessary for a police officer to put their life at risk to stop that when there are options that don't expose the officer to risk?

      Police work is not supposed to be a "fair fight". They are hired and trained to enforce the law and protect the public. They are not hired with the intent of making them cannon fodder.

      If a police sharpshooter happened to be in a position where they could have disabled Stephen Paddock from afar

      • Re: (Score:3, Insightful)

        by gweihir ( 88907 )

        You deliberately misunderstand my argument. Nice. Makes you one of the psychos and part of the problem.

        You should look up "asymmetric conflict" some time and then you will find that stupidity like this will cause more death, on both sides. One effect is, for example, that much more people are willing to surrender to a human that talks to them and exposes itself at least to some risk by showing courage. Conversely, if they can expect a "shoot first" policy (and these machines certainly go in that direction a

        • by uncqual ( 836337 )

          Having a tool available does not mean it is used indiscriminately. That is a belief that people have if they don't understand impulse control. I've keep fire extinguishers around -- guess what, I've never used them because the situation never called for it.

          Your dystopian view of people is not one I share.

          Police don't "owe" a criminal who is trying to kill them and others anything. There is no reason for police to take additional risk to stop, gently, someone who is doing so. Police should take risks to save

          • Comment removed based on user account deletion
          • Comment removed based on user account deletion
            • by uncqual ( 836337 )

              Your choice. If you feel lucky, fine.

              I've never used my seat belts for their protective purpose, never used my fire extinguishers for their protective purpose, never used my smoke or CO detectors for their protective purpose, and never used my homeowner's insurance. I also "probably" won't ever use any of them. However, I still have, maintain, and "use" them (in the sense of putting my seat belt on every time I've gotten in a car that was being driven by myself or others for the past 50 years).

              However, I su

    • I am happy to protect the police, and make their job less risky. That is a good thing.

      I just don't see this robot as the right way to do that. There are better options.

    • Killing people must never be easy and must never be risk-free or it becomes far too easy to do.

      Thanks for a new argument against death penalty.

      • by gweihir ( 88907 )

        Indeed. Ever wondered why the US kills so many people "lawfully"? Simple. Not even gross misconduct of the law results in any problems for the "lawful" killers. It is murder without consequences.

        I would say sure, if you really want a death penalty, have one. But if you kill an innocent, the jury, prosecutor, judge or police that has faked evidence or tortured a confession out of the accused, go on death-row as well. Without that it is no better than murder.

  • robocop does not need an warrant or Miranda.
    as it is not an person and that non person entity can not be called to the stand in court so you can't ask questions to it.

    • by uncqual ( 836337 )

      First, there are no requirements that a suspect be read their Miranda rights until they are both in custody AND being questioned. The definitely are not required before an LEO takes action to stop a imminent threat to herself or others. For example if you're arrested, as long as they police don't ask you any questions, they never need to "read you" your Miranda rights.

      In fact, you could conceivably end up convicted and in prison for life without ever having been informed of your Miranda rights if the case a

    • by Sloppy ( 14984 )

      It seems to be a contradiction, doesn't it? In Robocop 2 a company man explains to him that he's a product, not a person. But in Robocop 1 he says "my memory is admissible as evidence." How do we resolve this?

      But anyway, if I were Robocop and I wanted to maximize my chances of conviction (i.e. minimize the risk of the case being thrown out), I would read Miranda, just in case. What's the downside?

  • How much of a coward do you have to be to be completely incapable to subduing a cornered suspect without killing them?

  • by physicsphairy ( 720718 ) on Wednesday November 23, 2022 @09:10PM (#63075852)

    The whole point of using a robot in police situations is to engage non-lethally. Because you are not risking an additional human life that needs to defend itself. But the other part is that you can send the robots into high-intensity situations such as hostage negotiations without escalating what is happening on scene. As soon as robots become killers that will no longer be true, and you will give up the valuable intel and opportunity to talk-down the hostile individual through a remote presence.

    Even without considering the slippery slope, doesn't seem worth it if your primary objective is to protect civilians.

    • Even without considering the slippery slope, doesn't seem worth it if your primary objective is to protect civilians.

      The primary objective with this project is to protect the police. Literally.

      If the bot just so happens to get the 'bad guy' and not kill too many innocent bystanders, well chalk up another thumbs up on the bot budget increase for next year. All without putting a human officer at risk.

      And if it happens to fail, oh well. That's why you 'field test'.

  • better to ask for forgiveness

  • by presearch ( 214913 ) on Wednesday November 23, 2022 @09:34PM (#63075914)

    ..covered in blood.

  • I wonder how long it will take before the criminals begin using robotics / drones / remote devices to counter the police use of the same.

    Going to roll a lethally armed robot in on me ? Fantastic. Have an IED armed drone in response. ( Or as a pre-emptive strike since you know it's coming )

    • Criminals have been using remote controlled weapons for ages. Whether you fire a gun/bomb directly, from 100 m away in a foxhole, from 1000 m away in an overlooking building in Kandahar, or from 500 km away in the Kremlin, makes no difference to the outcome for the baby in the maternity ward in the target cross-hairs.
  • Look, this is a very bad thing. The "War on Terror" cranked up to 11 the militarization process started by the equally idiotic "War on Drugs", and it needs to stop.

    However, contrary to what the article is implying, the SFPD are not proposing some sort of AI killbot. They want to use their existing "remotely controlled, unmanned" robots as lethal weapons. Short of remote controlled bombs, there aren't a great many things one can do with these machines that can't be done with a human officer. That human of
  • I vividly remember that night down the road. Some lunatic downtown had just killed 5+ cops sniper style. The chief was given the option of sending more men into harms way, or they could deploy a makeahift weaponized bombsquad robot. I imagine Mr. Sniper had no idea this friendly little bot was about to light him up with some plastic explosives.

    The only reason this outcome was acceptable was because of the damage this guy had done. I also think this plan worked because the guy wasn't expecting to be blown up

  • Because a militarized police force sure worked wonders in the USA. Especially the poor/non white neighborhoods, right?

    Maybe we should train police to not be so gung-ho Rambo Commando when the situation does not warrant it.

    Defund the police? How about retrain the police?

  • ... "you have 30 seconds to comply!"
  • Many police forces around the world manage to detain suspects - even armed suspects, sometimes without even having access to guns, and only rarely using deadly force ...

  • If I were running the company that made these, I'd want an absolutely air-tight liability waiver stating that the company and anyone who works for it are not responsible of something goes wrong.

  • If you're going to grant that kind of authority to a group - the power to kill without risk - then I would think it would have to be a group that proved themselves capable of restraint in the absence of it.

    SFPD might have a problem there.

  • Who'll develop the AI for there to be a death penalty for pissing off a robot ?
  • Comment removed based on user account deletion
  • The SFPD will be announcing their first model at an event next week. They will be demonstrating the ED209.

Understanding is always the understanding of a smaller problem in relation to a bigger problem. -- P.D. Ouspensky

Working...