Hacker News Reader

Comment by slg | Hacker News Reader

>Because they oppose the drone warfare program, which is killing noncombatants in nominally allied countries.

I don't understand this logic. Drones are a weapon. They are not any more evil than any other weapon that the military uses. They aren't killing people autonomously. There is a human pilot behind every death. The pilot just happens to be hundreds or thousands miles away from the aircraft. Would we be fine with the military killing the exact same people if the pilot was simply on the aircraft that launched the missiles or bombs?

The distinct advantage of drones is to safeguard the lives of military personnel by not putting them in harms way. You can make an argument that this fact makes using military force politically easier, but I think the blame for that is once again on the military/political leadership and not the drones themselves.

If your problem is with killing people, it should be with the people who ordered the killing and not the specific weapon that was used to carry out the killing.

View on HN

"If your problem is with killing people, it should be with the people who ordered the killing and not the specific weapon that was used to carry out the killing."

Drone strikes, as they are used, are a more evil way of taking out identified targets of other nationalities. I find the concept of a nation state summarily assassinating people a bit troubling, but drone strikes make it even worse by making it imprecise and cheap.

Drone strikes without on the ground operators create huge collateral damage [0].

When operating in a civilian area, it's far less precise to handle the targeting based on a video stream and to send in a hellfire missile rather than have on the ground operators identify the target and pull the trigger.

Effectively, they are using civilian casualties as credits to buy military lives and convenient and rapid force projection capabilities. Any calculus that considers innocent lives as accepted collateral - when you have other options that are just more expensive or difficult can be considered evil.

[0] https://en.wikipedia.org/wiki/Civilian_casualties_from_U.S._...

Do you think that if part of Google's work were to make it more precise, would that be less evil?

If they could make sure they are killing only the guy they wanted, and act on that using pinpoint precision?

Uh, actually, now that you mentioned it, that would be even more scary.

That would be totally consequence free assassination device. At least the civilian casualties put some lid on the tendency to wantonly fly around in glorified RC planes murdering people.

The way I see it, if you are intending to assassinate a person (i.e. kill them not part of a military engagement) you could at least have the decency to have operators on the ground, with visual range to the target to pull the trigger. That's skin in the game. That's proper balancing of the calculus.

When you remove the chance of your own guys getting hurt, you are lowering the threshold to apply deadly force.

The whole point of due process and so on is to guarantee human rights for all and more importantly, put checks and balances on the actions of the political leaders.

When it suddenly becomes "ok" to kill people in automated fashion around the globe, that's a serious threat to all civil liberties regardless if one is a US citizen or not.

I think a free and open society is the only reason US has any credibility in claiming a moral high ground and things like eliminating targets on the disposition matrix [0] without due process erodes that high ground quite effectively.

[0] https://en.wikipedia.org/wiki/Disposition_Matrix

Every military technology from bows and arrows to GPS is doing exactly what you are objecting to by lowering the danger to the attacker and increasing the success of a potential attack. Drones are an incremental step in that process. It just seems strange to me to draw the specific line at drones when we are seemingly fine with other advancements that are almost identical in effect like stealth technology. Once again, my argument is not that drones are good. My argument is that drones are not inherently more evil than other weapons.

I completely agree with your last three paragraphs. But once again, none of that is specific to drones. It is true whether the trigger is pulled by special forces soldier from feet away, from an manned aircraft thousands of feet in the air, or a pilot in trailer hundreds of miles away.

I accept a global superpower sometimes assassinates people, and perhaps sometimes for good reasons.

My main opposition is to the factory like efficiency in which these extra-judicial executions are carried out by the drone pilots in a cost effective manner.

It gives me associations with other human organizations in the past which have carried out their mission with cruel detachment and efficiency.

I can't really put my finger on it, but it's just something I feel Stalin or Himmler or Robespierre would have totally agreed with.

Automated elimination of the enemies of the state? Wonderfull!

I used to feel similarly about guided missiles during Operation Allied Force (when the US inadvertantly struck the Chinese embassy in Yugoslavia).

My opinion on the topic has not translated into a reduction in automated warfare as a policy. On the plus side, independent of my opinion, technology has moved forward so that given similar circumstances, the embassy would at least have a chance that a drone pilot would realize during visual confirmation that the target has a big Chinese flag out front.

> Drone strikes, as they are used, are a more evil way of taking out identified targets of other nationalities. I find the concept of a nation state summarily assassinating people a bit troubling, but drone strikes make it even worse by making it imprecise and cheap.

There's no significant difference between a drone-strike and a conventional airstrike by a manned aircraft, at lest when it comes to targeting, precision, etc.

Drone-strikes might even be better than conventional airstrikes because the pilot may have less workload (therefore more focus) during the strike part, meaning fewer errors and mistakes.

> There's no significant difference between a drone-strike and a conventional airstrike by a manned aircraft, at lest when it comes to targeting, precision, etc.

But there's a significant difference in the sense that drones are the only way these strikes are conducted. Manned strikes would be more risky, since they expose a pilot, and less practical because manned aircraft can't be "parked" over a target for anywhere near as long as a drone. And whatever the legality of the situation, the political reality is that manned strikes look substantially more like an act of war.

So the difference is that many of these strikes wouldn't happen without drones. It's not about how the bomb is delivered, it's about whether we're assassinating people without a declaration of war in the first place.

Sounds like your issue is with the current political system not with drones. Direct your anger to the appropriate places.

I'm not angry at drones, any more than someone who opposed the V2 bombing of London was angry at rockets.

I have been against politicians who support this, though every presidential nominee for at least 12 years has supported it. But political opposition isn't limited to the voting booth. If a horrible act can be blocked by means reduction then dissuading people from enabling the means is a way to change politics.

Drones have the ability to be more precise than a typical forward air controller plus manned airplane as a drone can be above a target for much longer and that long duration video surveillance gives the opportunity to find the best time for a strike and to verify that it is the target.

You claim that a person on the ground is going to get better identification but they need to be hidden rather far away so that they aren't noticed and aren't hit. This reduces the typically achievable accuracy substantially from the theoretical maximum. Manned aircraft can't stay on target because of refueling needs, speed, and how loud they are, so you have very narrow windows of opportunity. Most ordnance from manned aircraft are also much larger than those used by drones, so manned aircraft typically have larger blast radius and can kill many more unintended people.

So while their can be narrow specific situations where FAC + F35 can deliver more precision, overall a drone provides much more precision and less danger to others.

This is obviously an emotional situation and many people don't believe that it is valid to kill specific individuals, as many others have claimed on this thread.

Sorry, I was imprecise. I did not mean it would be better to use conventional aircraft. I meant to claim that it would be better to use a human with a gun to do the assassination - at least when operating in an area with potential for civilian casualties. Use of heavy ordnance in an area with civilians should have no excuses if there is the alternative to use human operators.

I don't care how far up the rank of Isis or Al Kaida the person targeted is. Risking civilian casualties should not be acceptable.

A combat in a war and an assassination are two different things - I was condidering an assassination which by nature should be fine grained.

So the people who build the Death Star are innocent? The guilty don’t feel guilty, they learn not to.

I wouldn’t personally want to work in weapons development, but I certainly wouldn’t want to work in weapons development for a military that as a result of its offensive use of its weapons is making children in the third world afraid of blue skies.

But I have a gmail and I use google maps, so I’m not really in a position to judge.

Death Stars don't kill people. People kill people.

... ideally with Death Stars, millions at a time :D

I know you are being a little facetious with the Death Star line, but I think that bares a distinction because it would be a WMD and part of its intended purpose is to kill civilians. Likewise in our world developing indiscriminate killing WMDs like bio-weapons deserves extra scrutiny. Drones are not indiscriminate killing machines any more than manned aircraft. If they are used indiscriminately, the problem should be with the leadership ordering their use in that manner.

My overall point is that drones are not a unique class of weapon. You should either have a problem with Google working on a weapons program or not. Working on drones is no worse than working on manned aircraft.

Drones can be produced and deployed in higher quantities than aircraft, and with AI you won't necessarily need one man behind each drone. So yes, it is potentially worse than aircraft, for the same reason a police/army of bots could potentially be much worse than a "conventional" one.

>Drones can be produced and deployed in higher quantities than aircraft

Once again that is a political issue. If someone finds a way to suddenly reduce the cost of bullets by 50%, the response from the military shouldn’t be to buy twice the number of bullets.

>with AI you won't necessarily need one man behind each drone

The article denies that is how the AI would be used.

Just a matter of time.

It’s “purpose” is to protect the empire - it is a tool.

It’s intended as a deterrent.

Repeat indefinitely for all “tools”

(There’s an entire subreddit dedicated to it. The empire did nothing wrong)

You can Always justify making a bigger better gun or “tool”.

Don’t make tools you wouldn’t want used against you.

Tech engineers in particular are more than normally exposed to the ideas of weapons and responsibility of creators for their creations.

Dodging that is the opposite of being a coder/nerd.

Or, perhaps, the optimal survival strategy of a nation-state is to make the thing you don't want used against you until it becomes so onerous that international standards move to ban it.

Gas warfare was banned after WWI, but during WWI, both allies and central powers used it before it was banned. Nuclear weapons are banned... After the Allies used them twice.

International bans on intolerable weapons are a great idea. Unilateral disarmament of a useful weapon with no international ban may be suicidal.

Regrettably, in the world as it is, we need--extraordinarily--competent people working on our weapons of mass destruction, too (in particular, nuclear command and control). Not a job I'd want, and not one I'm happy exists, but one I'm sure grateful someone is doing.

> The guilty don’t feel guilty, they learn not to.

Along with the usual "If I didn't do it, someone else would".

Also known as 'the ultimate cop-out'

Drones are a worse weapon than a man on the ground with a gun, because they have asymmetric risks. The country with the drones doesn't risk civilian backlash from casualties. They get to start wars without much political pushback. And that means they can start wars for less justifiable reasons.

Mostly curious but do you feel the same way about body armor? Body armor reduces casualties as well and makes war seem less risky?

"We kill based on metadata"

Phonenumbers are the target identifier.

It is even looser than that, activity profiles and a graph of agents that fit some shakes marker.

They literally don't know who they are killing, only people engaging in certain patterns of behavior.

The whole thing is just a vaguely statistical murder game.

I had completely forgotten about that quote. Thanks for the reminder.

> There is a human pilot behind every death.

Google's work will now ensure that there won't be. Even with a 100% sure target identifier, it will now correctly hit the right target just as much as it correctly demonetizes youtube videos.

> Would we be fine with the military killing the exact same people if the pilot was simply on the aircraft that launched the missiles or bombs?

No.

> If your problem is with killing people, it should be with the people who ordered the killing and not the specific weapon that was used to carry out the killing.

In the context of this discussion the problem would be with the weapons designers.

That said, to imply drones are just like previous weapons is naive. The ethics of warfare change over time as the weapons used to wage it change. And death in war being even more abstracted away via drones (a process began with the first thrown rock and continued up through the carpet bombing of WWII and Vietnam) is certainly worth discussing.

If nothin else drones raise the question of scale - they allow for a lot more targeted killings.

Until the day war becomes an ugly memory we should all do as much as possible to examine and understand the forms of dehumanization it produces.

And if you disagree with the people ordering the killing, you might not want to fulfill their wishes for new weapons improvements, knowing that they'll use them for killings you disagree with. Or if it makes killings politically easier for them, then it's in your interest to not do that. Why is that unlogical?

Let's say your neighbor starts poisoning all the pets in your neighborhood, would you make poison for them because it's not the poison doing the killing?

What if the one that ordered the killing is an AI system designed by Google? Who is to blame? Remember that clients don't know the full implications when they ask for a service.

> I don't understand this logic. Drones are a weapon. They are not any more evil than any other weapon that the military uses. They aren't killing people autonomously. There is a human pilot behind every death. The pilot just happens to be hundreds or thousands miles away from the aircraft. Would we be fine with the military killing the exact same people if the pilot was simply on the aircraft that launched the missiles or bombs?

Not "fine", but I think "maybe a little bit less horrified" might be accurate, because a remote video feed apparently makes it even harder to tell what you're shooting [1]:

What the public needs to understand is that the video provided by a drone is not usually clear enough to detect someone carrying a weapon, even on a crystal-clear day with limited cloud and perfect light. This makes it incredibly difficult for the best analysts to identify if someone has weapons for sure. One example comes to mind: "The feed is so pixelated, what if it's a shovel, and not a weapon?" I felt this confusion constantly, as did my fellow UAV analysts. We always wonder if we killed the right people, if we endangered the wrong people, if we destroyed an innocent civilian's life all because of a bad image or angle.

Another former operator said similarly [2]:

Bryant stared at the screen, frozen. “There’s this giant flash, and all of a sudden there’s no person there.” He looked over at the pilot and asked, “Did that look like a child to you?” They typed a chat message to their screener, an intelligence observer who was watching the shot from “somewhere in the world”—maybe Bagram, maybe the Pentagon, Bryant had no idea—asking if a child had just run directly into the path of their shot.

“And he says, ‘Per the review, it’s a dog.’ “

Bryant and the pilot replayed the shot, recorded on eight-millimeter tape. They watched it over and over, the figure darting around the corner. Bryant was certain it wasn’t a dog.

Though in any case it seems like quite an awful argument that distancing yourself from whom your killing should be fine because killing-isn't-okay-anyway-so-whatever.

[1] https://www.theguardian.com/commentisfree/2013/dec/29/drones...

[2] https://www.telegraph.co.uk/news/worldnews/northamerica/usa/...

>”Would we be fine with the military killing the exact same people if the pilot was simply on the aircraft that launched the missiles or bombs?”

No, that wouldn’t matter obviously. Drones are only relevant because that’s what they are using: if it were helecopters and google were being asked to make computer-vision enhanced helecopter dashboards to make the program more efficient and more lethal people would have similar concerns.

This program in particular is troublesome because they have decided since it is out of self defense[0] it can be done outside of a congressionally declared war without public oversight[1] and to even target American citizens if the executive branch alone secretly determines they are linked to Al-Qaeda or another terrorist organization[2]. It has also been documented hitting first responders to ensure a target is not rescued [3] which would be called a war crime if anyone else did it.

Project Maven expands this new (in my view, extralegal) kind of warfare by making it more lethal and more automated. It scales this program up in a way that, say, making bullet proof vests for soilders or even tanks does not, because those are used in old fashioned congresssionally declared wars.

[0] https://www.justsecurity.org/wp-content/uploads/2016/04/Egan...

[1] https://www-m.cnn.com/2012/08/15/opinion/oconnell-targeted-k...

[2] http://investigations.nbcnews.com/_news/2013/02/04/16843014-...

[3] https://www.theguardian.com/commentisfree/2012/aug/20/us-dro...

In the immortal words of Rob Delaney, “Guns don’t kill people. People who say “Guns don’t kill people” kill people. With guns.”

/s/guns/drones

"there is a human pilot behind every death."

That statement won't last long term.

No, it'll become "there's a Google programmer behind every death".

The problem is that Google wants to give the military better weapons to kill people, knowing full well that they're intended to be better at killing people, and they're using the data we voluntarily gave them under the illusion that they wouldn't do evil things with it.

Evil things like making better weapons for people who kill innocent civilians.

> They aren't killing people autonomously. There is a human pilot behind every death.

Oh, so we don't need these weaponized AI developements then? But I suspect that with the AI you could more easily escape responsibility by blaming the software. Who faces the consequences when an autonomous drone bombs a school? "Oops, sorry, it was a bug."

More than the blame, it will dehumanise those deaths. Like a tree falling in a forest....

Weapons are not all equal. Chemicals can be used as weapons. Most of society generally agrees not to participate in chemical warfare. Some weapons are more evil than others.

> Most of society generally agrees not to participate in chemical warfare.

because of uncontrollable collateral damage. Same with mines that can't be deactivated.

Drones, on the other hand, promises precise strike, with little to no collateral damage.

If killing of "enemy combatants" can't be stopped, why not make it as precise and accurate as possible, and prevent collateral damage?

You do realize that drones are not precise at all? Plenty of collateral damage from them.

That is they are precise in targeting a specific location but they blow up everyone at that location?

Let's not even talk about double tap drone attacks. The second drone kills everyone attempting to help the wounded. Talk about do no evil...

Reply on news.ycombinator.com