Don’t Let Police Arm Autonomous or Remote-Controlled Robots and Drones | Hacker News Reader

Don’t Let Police Arm Autonomous or Remote-Controlled Robots and Drones

eff.org

282 pointsDiabloD3140 comments

The militarization of the police in the US is a huge problem that is out of control. I was extremely disappointed that Obama didn’t not only rein in this militarization but effectively made it worse. We are now having to deal with a generation of cops who think they are GI Joe but don’t have the training. This is why so many unarmed people are getting shot by the cops, because they start their day thinking they are in Fallujah instead of Cleveland.

I don’t understand why normal community police are armed at all. There’s absolutely no need for it.

It's a sad fact in the USA that you might encounter an armed criminal almost anywhere. I have personally seen a bunch of crimes involving gunfire, rather than mere brandishing of weapons.

On the other hand, police carry a lot of weapons, typically a handgun, pepper spray, a taser, a baton, and a heavy switchblade knife, plus a shotgun or rifle in the car. But they're relatively poorly trained and (even taking police claims at face value) deaths of civilians have resulted from confusion involving use of the wrong weapon.

Another factor is that urban areas are often policed by people who live in surrounding suburban towns, and end up feeling like they're going to work in a war zone every day because they are not embedded in the communities they ostensibly serve.

The vast majority of police never actually need their weapons. So why are they routinely carrying them?

https://www.pewresearch.org/fact-tank/2017/02/08/a-closer-lo...

There's tons of dangerous things we could carry around just in case, but we don't because they'd be seen as aggressive.

Same reason I have 3 fire extinguishers although I've never had a fire situation; you don't want to find yourself without if you suddenly need it.

Aer they over-armed, poorly trained, and often unreasonably paranoid? Yes. Being armed doesn't defend one against being surprised, and radios have saved far more police lives than guns on hips. But the work is by nature unpredictable, and the US is an unusually heavily armed society.

You're going to have a hard time recruiting or getting people to pay taxes for unarmed security. That gap is filled to some extent by community volunteers, but sadly that's not a very effective response in political terms.

This same argument could ostensibly be extended to anyone. There are armed criminals everywhere, why aren’t you carrying ${armory} on you at all times?

It seems to me like a police station, as well as select trained officers could be stocked with lethal weapons, while regular patrol officers carry nonlethals. Cops generally do not need a gun to pull someone over for speeding. If they are pursing someone known to be armed, it makes sense to do so.

Much like I don’t carry my laptop at all times, there are appropriate tools for appropriate situations and I’d like to see that extended to police.

>>This same argument could ostensibly be extended to anyone. There are armed criminals everywhere, why aren’t you carrying ${armory} on you at all times?

Many people do. I regularly see people walking around open carrying firearms in my rural town in WA state. People will have gun on their hip walking around Walmart or getting coffee at here.

I would like this too as I do not enjoy living in a heavily policed society, not to mention all the racial/class bias that seems to pervade policing in the US. But like I said, I think you'll have a very hard time recruiting or financing unarmed security as a public service in the foreseeable future. Investing more in alternative social services seems like the best approach for reducing demand, but you're probably looking at 5-10 years to produce statistically significant changes than you can rally voters behind.

To be clear, I'm not saying police should be heavily armed, but you have an uphill struggle persuading people who do think that, which includes many of the police themselves.

> This same argument could ostensibly be extended to anyone. There are armed criminals everywhere, why aren’t you carrying ${armory} on you at all times?

Some people do, and the rest of us have armed police a few minutes away. Also worth noting that in some states, legally carrying a gun is very difficult, such that lots of people practically can’t carry if they wanted to.

Disclaimer: I’m not a gun nut, nor do I have a desire to carry a gun.

> There are armed criminals everywhere, why aren’t you carrying ${armory} on you at all times?

I am, along with around 10% of adults in my state. Concealed carry by law-abiding citizens is more prevalent than ever, and I'm very glad this is the case. When I lived in the bay area, I was attacked several times a year. The only reason I wasn't crippled or killed was because I managed to outrun my attackers. Police were absolutely useless. Criminals knew that civilians weren't allowed to carry weapons, so they did whatever they wanted to anyone who was smaller, weaker, or alone. Now I live in a state with concealed carry permits and the only people who attack others unprovoked are the mentally ill. When a crazy person does this, there's a decent chance they'll be shot. This has happened at least once in the past year in my neighborhood. The crazy man died. The DA did not press charges as it was justifiable homicide.

Assuming you now live in PA (matches the 10% concealed stat you gave[0]), there's a much higher death rate by gun than CA[1].

FL and CO also have 10% of their populations holding concealed licenses and their gun death rates are worse than PA's.

The consensus is that guns reduce safety in communities[2].

[0] https://www.usconcealedcarry.com/blog/the-most-armed-states/

[1] https://www.americanprogress.org/issues/guns-crime/news/2019...

[2] https://www.scientificamerican.com/article/more-guns-do-not-...

I don’t live in Pennsylvania. I live in a blue state that has lower crime than California.

I don’t believe most of the studies about guns and violence. The people doing them are almost always ideologically motivated. (Including the “guns reduce crime” studies.) If you look at the details, you find them counting suicides as “gun violence” to boost their stats. Your second link makes that “mistake”. Another common mistake is to ignore all confounders. Yes people who own guns might be more likely to be murdered. Perhaps they own guns precisely because their fears are justified. Lastly, studies almost always conflate legal and illegal gun ownership. There is a huge difference in risk between a law abiding citizen who gets a concealed carry permit and a gang member who illegally carries.

The sad truth is that nobody is doing good research on this topic. Moreover, studies like these are barking up the wrong tree. I’m certain that restricting some rights would lead to improved outcomes, but that doesn’t mean we should restrict freedom of speech or voting. So too for the right to defend oneself from violent criminals.

The second link clearly breaks down gun deaths into two separate charts for the leading causes of these deaths, homicide and suicide. Where does it make that mistake and why do you have the word "mistake" quoted?

You should read the last link, it addresses pretty much everything you bring up, including the handicaps placed on research, adjustment for gun suicides, and the conflations that might tilt the arguments either way.

Maybe some people performing these studies are doing so for ideological reasons, but there's a clear reason why the pro-gun lobby has made it increasingly difficult to do good research and have clear-minded debates over the topic and it is not ideological, or even logical, unless you understand they're only doing it to maximize their sales and de-regulate their industry.

Gun ownership, and use, just like your arguments, tend to center around emotions and compensating for insecurities than around science and research (and common sense imo). They also tend toward low empathy justifications, like writing off gun suicides as not real gun violence.

> Same reason I have 3 fire extinguishers

Do you carry 3 fire extinguishers on your hip?

Or do you have them nearby and reasonably available to you for rare times that you do need them?

right, "its a sad fact that gun nut propaganda prevades Merica mentality making them think everyone needs a gun"

> The vast majority of police never actually need their weapons. So why are they routinely carrying them?

Your link doesn't support this. How many police have drawn their weapons?

Police in the US seem to draw their weapons as a normal part of many interactions with the public, so I guess it's 100%.

But I have absolutely no idea why they do that - it seems bizarre and terrifying to me. I'm certain they don't need to be doing it.

If you're drawing your weapon every day but never actually needing to use it... maybe stop drawing it?

You can use a weapon without firing it. For example, in this video[1] a rioter stole an M4 carbine from a police car. A security guard pointed a pistol and ordered the rioter to drop the weapon. He did. No shots were fired and the police carbine was recovered. Had the security guard not had a gun, I doubt the situation would have ended as harmlessly as it did.

1. https://www.youtube.com/watch?v=aVAMUejexa0

And in other situations pointing a weapon escalates the problem, causing people to respond like their life is in danger - because it is!

Also that example is a funny one to use, because if the police didn't have carbines in their cars people wouldn't be able to steal them, so this whole situation would never had happened in the first place!

Yes, sometimes introducing a gun escalates a situation. Sometimes it de-escalates by forcing compliance.[1] Sometimes it saves people's lives by stopping a threat. My point is that every country on the planet allows some people to carry guns. Different places just draw the line at different levels of screening, training, and places and times in which one can carry a gun.

There's also an issue of path dependency. What works for one country won't work for others. We learned from the war on drugs that banning something doesn't make it go away. There are more guns in the US than people. A lot of people here don't want them to go away. Unless the federal government wants to fight a civil war, Americans will be armed for the foreseeable future.

Several times in my life, I've had men who were bigger and stronger than me use violence against me. When a man tried to stab me in Oakland, I really wish I'd had a gun instead of nothing. When a stranger in San Francisco pushed me to the ground and tried to beat me with his bare hands, I wish I'd had something more effective than pepper spray. There were other occasions that I don't want to get into. As I said in another comment, the police were absolutely useless. They never caught any of these people. I'm certain that after my encounters with them, these violent criminals victimized dozens of others. These things happen so often that many people don't even report them to the cops. Why waste more of your time when you know nothing will come of it? The statistics in the SF bay area are definitely underestimating the magnitude of the problem.

I've seen similar crimes happen to others. A few years ago I watched a man try to kidnap a woman at knifepoint. The woman was powerless and so was I. The only reason he was stopped was because men with guns showed up and ordered the man to drop the knife. He decided he'd rather be in prison than six feet underground.

Now I live in a place that lets law-abiding citizens carry firearms for self-defense (after background checks, training, and fingerprinting). The difference in people's demeanor is night and day. Civilians aren't scared here. Homeless people exist, but they're not aggressive like they are in SF and Oakland. After moving, my girlfriend (a California native and lifelong Democrat) was so shocked by this that she revised her position on guns. She's now learned to shoot and is interested in getting her own concealed carry permit.

1. https://www.reddit.com/r/PublicFreakout/comments/onpu1m/conc... Here's a News article explaining what happened: https://www.fox6now.com/news/demanded-a-refund-man-charged-a...

Training isn't the issue, it's the lack of accountability. Military members are under the threat of being court martialed and spending time in military prison. They know they have less rights in military court than they would as civilians, and that punishment can be harsh.

Cops, on the other hand, know that the system will bend over backwards to accommodate whatever transgression or crime they commit. They can and do act with impunity, because they're actively aware of that impunity. They know that if they get caught, in the worst case scenario, they'll get a paid vacation, their boss will allow them to resign, and they'll have to work one town over.

I'll second this. In the military (or at least in the Marine Corps, from my experience), the person to your left and right is not only there to help you, but to hold you accountable too. It's started early in training that you don't "let your buddy off the hook", you f** them up if they do something stupid. The goal is to uncover all the dirt and get it cleaned up, not to hide it. And your buddy will testify against you, because they know it's the right thing to do, and you'll spend years in the brigg. It's a hard culture, but it's built to be self-filtering, self-cleaning. That's where the idea of honor comes from.

From everything that I've witnessed and heard, police culture is the polar opposite. You "do favors", "hook each other up", and "overlook mistakes". All of this breeds the bacteria, rather than killing it off.

There is money (and usually in cash) in the police line of work. It's much harder to make any money from the public in the army. So there is much less at stake.

Police can make arbitrarily good amount of money if they are corrupt. That's what is unbalancing things.

Combat vets aren't the problem. They are less likely to shoot. It's the wannabes doing paramilitary cosplay who are the heart of the issue.

Statistically, combat veterans are 6 percentage points more likely to shoot than non-veterans. (32% versus 26% have discharged their service firearm while on duty, other than at a gun range or during training.)

https://www.pewresearch.org/fact-tank/2017/02/08/a-closer-lo...

That stat doesn't mean much to me though. I'd want to know about unjustified discharging of weapons. I'd imagine cities like LA, Baltimore, Detroit, etc are more likely to assign combat veterans to the places with higher crime, particularly gun violence, than anyone else on the force. I'd also expect former military to be involved in S.W.A.T. more than non-veterans. And for that reason I expect their likelihood of discharging a weapon to be higher.

\\ This is why so many unarmed people are getting shot by the cop

How many unarmed people are killed by the police each year in the us? When I ask this question, most people guess "thousands" or "tens of thousands". The real number is around 50.

Of course the ideal number is 0 but the US is very far from having a " police killing unarmed people " problem.

Who decides if the person killed is unarmed again?

That's an interesting point. So how did we decide whether there a real problem or not?

Source?

This article says it was almost 1000 last year. https://www.washingtonpost.com/graphics/investigations/polic...

A few years ago police shot a man neareby when they mistook a water hose nozzle he was holding for a handgun.

That’s total, not unarmed.

My brother is doing research on suspicious usage of "Suicide by cop" as a cause of death, a phenomenon where police departments and coroners may be overly broad in their application of the label. I hadn't thought about this, but apparently this imaginative labelling _may_ be hiding unfavourable statistics .

Although I doubt this would change the number from 50 to 10,000.

The Washington Post says the police killed 413 unarmed people from 1/1/2015 through 7/16/2021, about 63/year.

https://www.washingtonpost.com/graphics/investigations/polic...

They're about 30 years too late on this one. Police and the FBI have been using EOD robots with shotguns for a long time now and have even killed a few people with them. In a hostage situation or standoff, they'll send in the robot with food or a phone hanging from the arm, which has a Remington semi-automatic shotgun attached to the base. Ruby Ridge is one of the more famous examples.

If you're referring to the 1991 Idaho - Ruby Ridge then no, robotics were not deployed.

Weaver's testimony to Congress:

"Come out and get the telephone, and later on I got a chance to look at that robot sitting out there, and it had a sawed-off 12-gauge shotgun on the side of it, bolted to it, aimed aright at the telephone, which later they said was empty. I do not believe that. I will never believe that. If they want to negotiate with you and they honestly want you to come out and pick up a telephone, they are not going to have a shotgun there, because they know if I see that I am not going to come out after that telephone. They was hoping I did not see it, and the way it was aimed at the house, I did not see it the first time I looked at it. The second time I looked at it, I saw the hole of that barrel pointing up, and I got to looking better, and I said, holy cow, that is a sawed-off shotgun pointed at the telephone. I did not see it the first time."

The guy in Texas was the first IIRC killed with an explosive. https://www.cnet.com/news/dallas-shooter-killed-by-bomb-equi...

And that appears to be the first case on US soil where a robot was actually used to kill.

It's not. Police agencies have taken care to avoid headlines that might raise unpleasant questions about killing suspects with military robots, so it's not something that's easily googleable.

Are there any other examples?

I don't deny that they could be quiet about it - but it actually seems easily knowable given the reports that finally come out.

Plus - there are so few situations where the suspect has barricaded themselves anyway. Its not like car chase suspect or people running away from cops have been able (in the past) to be attacked by anything other than cops guns or tasers or guns.

Most news stories from the late 80's/early 90's don't seem to be available outside of paid archives, so I don't have any links to provide.

PBS on Ruby Ridge:

"The Weavers wouldn’t go near the robot — even after they took off the shotgun that had been strapped to the front of it."

https://www.pbs.org/wgbh/americanexperience/features/ruby-ri...

The primary purpose of that shotgun is IED/UXO disruption, ostensibly. It's unfortunate that the robots have been used as weapons, and that people aren't in prison for it. But I think arming AI-equipped drones specifically to kill is a much more worrisome threat.

I mean, I get that some jargon is clear to those who have an interest in the topic. But how the hell is it acceptable to use sth like "IED/UXO" on an article about ethics? This is so obviously not a well known term (and not searchable).

My mind boggles.

Explosive Ordnance Disposal (EOD), Improvised Explosive Device (IED), Unexploded Ordnance (UXO). Civilian police use of robotics started with EOD robots, often borrowed from the military.

I'm guessing you're not in the US (or UK) where IED and PTSD acronyms have been used so much in the context of physical and psychological injury in the Middle East tha they are pop culture hard to avoid knowing. Certainly, in the US, the first term that pops up in a search for IED is:

IED

/ˌīēˈdē/

noun: a simple bomb made and used by unofficial or unauthorized forces.

Danger UXB (UneXplodedBomb) was a popular British drama 30+ years ago.

I assume you aren't American, Canadian, or Western European. The news constantly discussed IEDs in the context of the Iraq and Afghan wars and UXO is used whenever a bomb is discovered leftover from one of the World Wars.

It is also very searchable, at least to me. The results returned are very relevant.

https://www.google.com/search?q=IED%2FUXO&rlz=1C1CHBF_enCA85...

>But how the hell is it acceptable to use sth like "IED/UXO" on an article about ethics?

how the hell do you expect to understand an article on autonomous drones ethics without knowing that IED/UXO means? It makes obvious that you lack basic knowledge of modern factual landscape in the domain, and how do you expect to understand ethics of the domain without basic factual knowledge of the domain?

The argument that those acronyms are a requirement for the understanding of the ethics of autonomous killer robots is a very difficult one.

it is a major drone use case. Understanding an ethics of technology without knowledge of major use cases - that is a difficult argument.

My bad, dude. The ambiguity has been addressed in other people's replies.

Better robots than people in that position.

Is that a joke?

Didn’t they use a robot to take out the Dallas sniper who was targeting whites and police. I don’t really see what is wrong with that situation. I get why there is fear about using them in the general case however.

Never too late to change how things are done.

> But police technologies malfunction all the time.

> When, inevitably, a robot unjustifiably injures or kills someone--who would be held responsible?

That Robocop scene where ED209 kills that poor executive used to be a joke. Though now that I think of it, Paul Verhoeven was probably serious.

https://www.youtube.com/watch?v=j5NAswnEewg

It is a perfect moment since you can see the low level people panicking while the true bosses are like "ugh, this is going be so annoying to fix", completely uncaring about murdering someone.

What a wonderful glimpse into corporate soul.

The Dallas police did that in 2016:

https://edition.cnn.com/2016/07/12/us/dallas-police-robot-c4...

I am not sure I agree with the EFF on this one.

If the future of cars is autonomous to save lives, then the future of police could be the same for the same reasons.

The 1951 version of the The Day the Earth Stood Still, with an all-powerful robot GORT that neutralized human aggression left an imprint on me. I honestly think a legion of GORTs will be better than a legion of humans - if only because I believe the end of the line for robots will be significantly better than human potential.

John Henry 'loses' against the machine in the end, now matter how much we root for him.

> If the future of cars is autonomous to save lives, then the future of police could be the same for the same reasons.

The presumption that there is such a thing as "the future of police" is worrisome. The now nearly 200-year experiment of designating a tiny subset of the population as exclusively responsible for public safety and law enforcement goes very poorly. The abolitionist movement is as strong today as any time since the conclusion of the American Civil War.

When do we start getting serious about visualizing a future without police?

The police abolitionism is much stronger in the US, where you can trace lineage of police to slavers and union-busting hit squads, than in the rest of the civilized world.

This is accurate. From Time's "How the U.S. Got Its Police Force"[1]:

> The first publicly funded, organized police force with officers on duty full-time was created in Boston in 1838. Boston was a large shipping commercial center, and businesses had been hiring people to protect their property and safeguard the transport of goods from the port of Boston to other places, says Potter. These merchants came up with a way to save money by transferring to the cost of maintaining a police force to citizens by arguing that it was for the “collective good.”

> In the South, however, the economics that drove the creation of police forces were centered not on the protection of shipping interests but on the preservation of the slavery system.*

Police were union-busting, as well:

> For example, businessmen in the late 19th century had both connections to politicians and an image of the kinds of people most likely to go on strike and disrupt their workforce. So it’s no coincidence that by the late 1880s, all major U.S. cities had police forces. Fears of labor-union organizers and of large waves of Catholic, Irish, Italian, German, and Eastern European immigrants, who looked and acted differently from the people who had dominated cities before, drove the call for the preservation of law and order, or at least the version of it promoted by dominant interests.

[1] https://time.com/4779112/police-history-origins/

Policing is far superior to what it replaced.

Previously it was the army, mob rule, or mafia rule (sometimes legally sanctioned like a Lord of an area).

It's 2021 - well into the age of the internet.

What relevance has the practices which police ostensibly supplanted to the matter of how we handle public safety post-abolition?

Compared to where we were it is an amazing success. There is a high standard for replacing it with something else.

Is the American model of policing better? Not really, they are instutionalized slavecatchers put in place when the south lost reconstruction to the KKK.

The British model has a chance, they at least have a moral foundation to work from. https://lawenforcementactionpartnership.org/peel-policing-pr...

> When do we start getting serious about visualizing a future without police?

There will always be rules, people will always break them and some group of specialists will always be trusted with extra responsibility to deal with those people.

But that doesn't have to look anything like what we have now. Arguably, courts meet your description.

When a man comes home and finds his wife in bed with another man whom he then murders, how will the courts find that man and bring them to justice?

This has never been the job of police, but of the much older institution of the sheriff (though today, there's hardly a distinction).

I'm clearly talking about today and the future.

Right. So, the most obvious answer is to simply restore the institution of the Sheriff while abolishing police.

An even better answer is to reimagine decentralized community security with all of the capabilities of an information age society - cameras, drones, etc.

What problems are you trying to solve? I think there were a lot more very serious issues with law enforcement the further in the past you go.

> An even better answer is to reimagine decentralized community security with all of the capabilities of an information age society - cameras, drones, etc.

You're going to have to be a _lot_ more detailed when you're proposing something so radical if you want me to at all understand how that would actually work and work better than what we've got now. What you just described sounds to me like my entire community invading my privacy.

Those who say they want a future without police are often told to "move to Somalia," or some such place where local warlords provide the only semblance of law.

Those who say they want to increase authoritarian influence are told the same thing.

This makes me think we are probably close to the optimum solution. The police just need to be held accountable for their actions. Right now, that seems to be the biggest weakness in the system. It's unrealistic to expect perfection from the police or from members of any other profession, but when things do go wrong, it takes negative feedback to correct it, and the police unions have proven exceptionally skilled at breaking the corrective feedback loop in multiple places.

> The police just need to be held accountable for their actions. Right now, that seems to be the biggest weakness in the system.

I worry that the lack of accountability might be fundamental to the current system, and the "just" in "just need to be held accountable" is not even remotely as trivial as it sounds.

Agreed. That's usually true when anyone including me uses "just" in a sentence like that.

> That’s true whether these mobile devices are remote controlled by a person or autonomously controlled by artificial intelligence, and whether the weapons are maximally lethal (like bullets) or less lethal (like tear gas)

Autonomous fire control, I agree. But remote fire control on non-lethal weapons?

The EFF’s argues a physically-present officer “will have better information about unfolding dangers and opportunities to de-escalate.” I want evidence. A remote officer has no threat to life. They can’t use a self defence argument in court. And they can think through things calmly. We also have more opportunity to record their inputs and actions and thus increase accountability. The “hackers will inevitably try to commandeer armed police robots” concern is real, but not insurmountable, and de-risked by limiting networked drones to non-lethal weapons. As for “capabilities of police to conduct crowd control by force” being “too great,” forcing humans into harm’s way seems like a wrong, circuitous way to address it.

Police brutality is real. But so is the brutality of criminal activity. We owe this question more thought than a knee-jerk NRA-esque reaction.

Crime, including violent crime, has been on a long downward trend since 1994.

> But remote fire control on non-lethal weapons?

Non-lethal weapons are used for compliance, and are often abused, sometimes with fatal consequences. Turning it into a video game for the police drone operators and detaching them from it will breed more abuse.

I particularly don't want to live in a world where can you get tased if you don't do what a drone tells you to do.

> Turning it into a video game for the police drone operators and detaching them from it will breed more abuse

Again, where is the evidence for this? We’re proposing a ban that puts people’s lives at risk. You’re willing to do that on a hunch?

Unfortunately the cake is already in the oven in this. There will almost certainly be major changes to civil liberties coming as a result of the “capital insurrection”.

If you want to fight against police overreach, you kindof have to fight against it even when it comes for your enemies.

You guys are way too late.

The police already have armed robots. Have already used them to kill. And police overreach is already a thing.

All of it happened decades before the capitol insurrection. I'm surprised so many seem unaware of this.

If the police can do it, so should private citizens. If not, then neither the police should do it.

This simple calculus needs to be the basis of police armament generally, not just robotics.

I’m less worried about the police doing this than I am rogue nation, groups, and individuals. It’s only a matter of time until Hamas launches an armed, low flying, autonomous drone swarm in the heart of Tel Aviv. The book “Kill Decision” by Daniel Suarez convinced me that we have less to fear from government and more to fear from private actors when it comes to lethal drones.

Actually, I’m really surprised it hasn’t happened already given the low cost.

"I'm less worried about the police doing this than I am rogue nation, groups, and individuals."

Governments have killed exponentially more of their own people throughout history in genocides, pogroms, civil wars, starvations, and so on than have ever been killed by rogue nations, non-government groups or individuals.

I always thought targeted assassination will be the primary use of autonomous weapons. And since I don't really see any obvious means of authoritative attribution, I'm not really sure how you would stop it? I think it will be ultra popular. Not only with cartels looking to rid themselves of annoyances in Mexico. Also with people looking to remove any mayor they don't like, or any police chief they don't like, or, more commonly, any boss they don't like, or even any neighbor they don't like. Most people in society would be almost a free target. Combine that with the fact that most people would have access to this tech and I could see tit-for-tat killings going on all over the place.

I'm really interested in seeing what law enforcement and military thinkers are working on to solve this problem. It really will be fascinating to watch this space, because I could see it getting out of hand on a time scale shorter than we'd at first anticipate.

I remember listening to a guy on Business Radio, who was operating in this space. His business model was basically making sure that drones can't be easily disabled, jammed and so on along with the other side of the coin where 'bad guy' ( I so hate that phrase ) drones are contained quickly.

That was two years ago. I assume this is part of the reason some drone regulations were implemented.

YOU may not be, but if I, or my loved ones lived in a high-patrol area known for police brutality, I certainly would be worried.

Drones should be much easier to intercept than rocket / artillery shell attacks which have become significantly less effective. https://en.wikipedia.org/wiki/Iron_Dome

This is not clear at all. Rockets and artillery shells follow high trajectory, extremely predictable ballistic paths. If your radar is good enough to accurately spot the rocket or shell against the entirely blank backdrop of sky, you can shoot it down. Drones, on the other hand, can fly low to the ground and move in unpredictable ways if needed so that 1) radar can't find them easily, and 2) even if it does, your counter weaponry will probably have trouble doing much.

A lot of anti-drone weapons today are based on the idea of disrupting drone-to-controller comms to force them to return or land, but if you setup your drone on a one-way trip with zero need to communicate to a controller, your countering options get limited pretty quickly.

Off the shelf drones are so slow they don’t need to be predicted. Signal jamming is the default because it’s safe, cheap, and effective. However at Phalanx CIWS is more than happy to shoot down a drone, but using one for anti drone work presents some inherent risks as it’s quite possibly going to target sea birds by mistake.

Anyway, if you’re thinking about a custom built drone which follows pre programmed flight paths to a preset target, congratulations that’s a cruse missile. Various versions have traded off speed to follow the terrain more closely, but speed is considered more useful than extremely low speeds.

Jamming GNSS is the next obvious choice. Low noise gyros and accelerometers needed for inertial guidance/dead reckoning over long distances are expensive.

Now something that could fly via visual guidance like a Skydio is interesting... Counter with giant spotlights or laser dazzlers to blind the onboard cameras?

Terrorist organizations have started drone attacks - https://www.livemint.com/news/india/jammu-airport-attack-fro...

Israel uses "AI" controlled drones with no human contact to terrorize people in Gaza. Palestinians absolutely already manufacturer and deploy simple drones against Israeli military targets. Israeli terrorist army force (in a euphemistic doublespeak manner called defense force) does way more damage against civilians while Palestinians resist against an occupier.

Hamas is already doing this: https://www.forbes.com/sites/davidhambling/2021/05/14/hamas-...

I'm only surprised we're not seeing more of this from domestic terrorist groups. Anyone with minimal technical skill could probably order a hobbyist drone, attach something nasty, and figure out how to remotely detonate it with a phone.

Also, I suspect that arguing cops shouldn't be able to use armed robots for extreme situations like hostage standoffs—when they already use robots for bomb disposal and reconnaissance, to say nothing of long-standing measures like deploying snipers on rooftops—is going to be a tough case to make. Might have more success trying to limit the situations where they can use armed robots—say, a hostage situation or ongoing terror attack, rather than shadowing a protest.

Using drones wasn't going to dramatically increase the damage someone like the Boston Marathon bomber did or decrease his chances of getting caught. Would be domestic terrorists are deterred by the fact that they are domestic and will eventually be arrested given the dragnet of internet and CCTV surveillance we all live under, not because they can't somehow fly a bomb into a stadium instead of leaving a pressure cooker backpack. Drone availability is way down on the list of things a normal person would consider before deciding to pick up domestic terrorism. Top of that list is probably whether they can win a shootout with the police force when a dozen swat vans inevitably show up at their address.

"I'm only surprised we're not seeing more of this from domestic terrorist groups. Anyone with minimal technical skill could probably order a hobbyist drone, attach something nasty, and figure out how to remotely detonate it with a phone."

Me too, it is not high tech at all, basic in doing a remote circuit and they now even have "farming" drones which you can replace the main chemical with whatever you'd want...

https://www.youtube.com/watch?v=Dn7Q4oIrP3I

> Hamas is already doing this

And the IDF too: https://www.thedailybeast.com/israel-is-sending-robots-with-...

I like this example, the Slaughterbots: https://www.youtube.com/watch?v=HipTO_7mUOw

Slaughterbots (from 2018) is pretty terrifying. Do you want your local police force to have this?

https://youtu.be/HipTO_7mUOw

resounding yes. With an armed drone/robot being in the line of perceived danger instead of a police the police looses that "split second decision" argument that has been letting them to kill all those people just because the shadow looked like a gun instead of a say giraffe.

You remove police from the harms ways - you remove the glorification and their high-caste "above the law" status, and their union wouldn't have such huge political power, nor public support for those state budget crushing humongous pensions.

Also with a robot shooting precisely there would be no need for that "stopping power" BS with illegal under Geneva hollow point bullets killing that violently that the police so likes to use (driving end result more toward death due to large blood loss and tremendous tissue damage inflicted by the hollow point - all unnecessary really). The robot(s) can just precisely disable arms and/or legs with much less severe wounds (like police in some other countries do) using smaller caliber or even just an electric shock or temporarily motor nerve blocking agent, sleeping chemicals, etc.

Also the robots will treat people equally, and thus big people getting the same law enforcement treatment as the rest of us little people may force the big people to adjust the law enforcement toward more reasonable approach. (i mean for example why the corporate IT support sucks so much? Well, what you would expect giving that the execs have their separate IT. Imagine if they had to use the same IT as rank-and-file and the IT couldn't tell whether it is an exec or regular Shmoe :)

What's the over/under on ED-209 making a comeback?

I was always wondering which one would become true first. Robocop or Terminator.

With enough abuse of technological advances, both are only a matter of time.

I hope humanity figures out a peaceful utopia where everyone can be fed with tablets which give all nutrients and are freely distributed, and robots do work so humans don't have to. But looking at human history, I remain sceptical, it looks like the robocop/terminator scenario is more likely. Throughout history, humans have thrived for power and wealth accumulation, by peaceful or violent means.

The day might come where many people will ask "how did we permit this to happen?".

It will likely happen because the politicians , generals etc will say "like this, we don't have to risk human lives".

They might be remote human operated at first, but the push for autonomous beta test will be inevitable.

One bug or vulnerability will make the American police violence today look like a picnic in comparison.

Maybe I am totally wrong, but this is a gut feeling.

I picture ED209 from Robocop 1. Just a matter of time I suppose.

It's our children I truly fear for- as in by the time they are in their 40's who knows how bad things will be.

Edit-added last stuffs.

How about we spend less time debating which horrifically powerful weapons to make or not, and more time on how to foster a society that doesn't have any use for them?

Five thousand years of recorded history suggests that's a pipe dream. Such cultures only exist in fiction (that we know of). Utopia is often described as imaginary.

Utopia literally translates to "no place."

What's your argument here? That working towards something perfect is inherently a waste of effort? It should not need to be pointed out that trying to change things for the better does not have to create a utopia to be beneficial. Reducing crime in non-militant ways (and therefore a reliance on police) does not need to remove crime to be a good thing to do.

I can reel off a list longer than recorded history of entirely unprecedented events of worldwide impact that have happened in the past few decades versus the past 5000.

The imagination being entertained here is your own; that humanity's future from this point could be even remotely similar to anything that's ever happened before.

Every choice we make as a whole now is of extreme importance, and almost nothing we can conceive of can be sensibly considered to be off the table.

Doesn't the solution to your second question probably involve settling the debate on not making weapons at all, anyway. Which is the first question.

This seems like a false equivalency between a given individual and the state: I can ask what weapons I'm allowed to have without asking what weapons the state is allowed to have.

No, that's the entire point of the statement. The debating of such things is counter to the stated goal weapons are purported to be brought into existence to achieve.

We don't stop people killing and harming each other by figuring out how to kill/harm them (or threaten to do so), by ever more powerful means.

We stop people killing each other by figuring out how to stop them (and us) wanting to do anything remotely like that in the first place.

Unless we change direction, we will end up where we're going.

A police department can't gain budget to not buy things though

Why is the police department the one working to make the police department obsolete? That seems like mistake number one if we're going to try to demilitarize the police.

Police only exist in a complex enough form of social consequence worth arguing about, where people don't have the material and mental means to comfortably enjoy life.

Concepts such as law, order and defence are topics that readily devolve - precisely because of widespread mental deficiency - into counter-productivity.

The base useful question we can ask as a species supposedly seeking peace, is how to best cultivate the material and mental means that demonstrably create it in us.

Here's the "problem": on a worldwide level there are bad actors who will for sure deploy armed robots. Period.

So, do we (we = any country in the world) want to be ahead of this technology or behind? Leading or lagging?

Because the leaders will have the power to overthrow the laggards, and if history has served us as a teacher, if it MIGH happen (capability) it WILL happen.

Add to the mix overpopulation and the related scarcity of resourced (e.g.: water, protein rich food) and now we have an interesting GLOBAL Game Theory case.

There must not be a mine shaft gap!

The police don't respond to foreign threats, they enforce laws domestically and locally. Armed police drones aren't going to be used against foreign threats, they'll be used against everyday US residents while enforcing domestic and local laws.

That's a legitimate debate, but you're exploring a military problem in a criminal context. True, gangs and drug cartels could and perhaps will use robots to gather intelligence or kill enemies, including police. But let's be realistic here, they could attack each other with faked up tanks or from stolen helicopters...but that almost never happens.

Aye, There's the rub - these bad actors who have already demonstrated the use of AI enhanced robots/drones that we must face a defeat or be beaten. Make no bones about it, there are latter day hitlers out there running countries like Iran, North Korea etc - much as we dislike the concept we can not allow them to defeat us.

How often have proposed limits on police power actually been effected (or effective)?

It seems that every year, year after year, and decade after decade police are only ever getting more and more powerful.

On the upside, three states so far have abolished qualified immunity (New Mexico joined the list earlier this year). The first was Colorado in 2020. So there's hope this may spread to other states in the coming years.

The problem is that abuse tends to vastly outpace incremental reform. We are just now making serious strides to legalizing cannabis and decriminalizing other sorts of drug use. That's great, but stack it up against the length of the drug war, the billions involved, and the number of lives it has impacted.

I think _autonomous_ weapon systems that pick targets and fire autonomously are obviously scary.

However - and this is probably not going to be popular - I think if robots are remote controlled, it's probably for the best if lawful-violence-monopoly-appliers (i.e. the state) uses them instead of putting humans out there with weapons.

I think a lot of situations escalate to higher echelons of violence when the (e.g.) police officer is faced with a threat to their life and needs to quickly go up the escalation of force ladder to disable the opponent, euphemism for shooting them. If instead a robot is the agent being threatened, the equation would be much different. There would be no need to kill the human opponent, since a robot is not worth sacrificing a human life for.

To me, remote controlled weapon systems are a great way to avoid unnecessary escalation of force. And escalation of force are, in my opinion, what is at the heart of most display of police violence (police cars driving into crowds, police officers shooting someone). Of course there's no denying that there are many cases where force was unjustifiably misused, but I'm an optimist at heart and I think the majority of cases isn't so, and that robotic telepresence would be a good way of defusing a lot of situations that would otherwise have escalated.

>> If instead a robot is the agent being threatened, the equation would be much different.

There is no need to have an armed robot then, since self defense is not necessary. It is IMHO very bad for police to shoot in domestic calls, as you never really know who the "bad guy" is, and killing someone is equivalent to saying "is someone is gonna kill some here, its gonna be me".

For other situations like robbery maybe it's not as bad, but again there's no need to protect the robot.

Not as big a need to protect the robot, but the robot might be trying to protect other people, and thus would require some method of disabling an attacker. Egregious example: a hostage situation, or an active shooter situation. With a remote controlled robot, you can use less lethal weapons against the attacker, since the robot can be sacrificed (whereas I assume we'd rather not sacrifice police officers).

And I put it to you that being safely miles away in an airconditioned box without risk of being swarmed by a mob for shooting first will make police more likely to escalate violence much sooner.

How are we supposed to know which is correct before it's too late?

Historically, providing alternative methods of de-escalating situations to law enforcement has proven effective in diminishing usage of lethal force. In the end it's speculation, and your guess or mine.

Your guess is premised on (1) the belief that police officers will act more carelessly, thinking they're in a video-game if they control remote machines. My guess is premised on (2) the belief that undue escalation of force happens when police officers feel their life is at risk, and removing this element from the situation will improve things.

(1) assumes that police officers are acting violently out of disregard for others, and (2) assumes that police officers are acting violently out of fear for themselves.

I'm a humanist and I think (2) is more likely than (1). It seems a simpler and more pragmatic assumption, to me, that police officers may act overly violently due to fear. I much prefer dealing with someone who's able to maintain a cool head, than dealing with someone who feels threatened.

with enough check and balances I think this is only a positive. It will lead to faster and more accurate police work.

It will also take the "life was in danger" aspect away from policing in which people are killed unncessarily. A drone can target a leg of a suspect without having anyones life in danger.

The police are not moral arbiters. They are policy enforcers. So you basically have a robot that has a soul and fears mortality going around right now. This is more dangerous than actually having a robot do the job.

Some police are community protectors but that has gone the way of the dodo bird a long time ago. Especially in big cities.

Sorry to be a downer but it will happen if it hasn't already, and the biggest reason is cost savings which is pretty hard to argue against when the budget is up for discussion at the local gov.

Having drones patrol large areas, running plates and face-matching against a warrant database is much quicker and cheaper than having people doing it. They're also cheaper than helicopters to follow fleeing suspects, and a bit harder to detect.

I agree though it's a huge privacy issue here, but when it comes down to money it'll be a tough fight.

> Having drones patrol large areas, running plates and face-matching against a warrant database is much quicker and cheaper than having people doing it.

Why would it be cheaper than metal poles with cameras attached to them? Surely stationary devices will be cheaper and safer.

I guess a drone can just land on a roof (or even street-level)?

Basically a battery powered camera that can be deployed very quickly to an area of interest.

I have no problem with any of this, and I'm a full-blown police abolitionist. I love drones and their potential for public safety assurances.

However, it's a completely different discussion when we talk about arming them.

We shouldn't allow anyone to do this, not just the police. Although it already happens, and the American military already does this internationally. Drone attacks flourished under Obama's administration basically, look up the incident around 2013 when US military drone fired at a wedding convoy in Yemen and killed 10+ people.

Drone attacks increased with every president. The one after Obama also set his own record and even tried to hide them from reports.

If the US doesn't do it the others will(Russia, China etc).

That's an empty argument. Criticizing American military and government doesn't mean you can't criticize others.

My point is that the enemy would gain a tech and tactical advantage if you restrict the usage of robotics in your army.

They already have the advantage, American military is stretched too thin because politicization think they're running an empire. The pentagon is a corrupt institute. They made triple-digit trillions in "accounting adjustments" over the years.

Hey, officer, send this guy to do drone job manually.

Reply on news.ycombinator.com