"SKYNET" A.I. Drone Program May Have Killed Thousands of Innocent Pakistanis

[Madmick]

https://www.theguardian.com/science...algorithm-really-killed-thousands-in-pakistan
http://arstechnica.co.uk/security/2...-may-be-killing-thousands-of-innocent-people/
First, yes, they actually named a military program with killer flying robots "SKYNET". They actually fucking did that.

Second, I'm too goddamned depressed right now to say much more in this OP. My country continues to fail me again, and again, and again, and again. I'm still seething from the UN hospital. Still bring that at up at dinner tables. This is not what we are supposed to be, what I want us to be. We are supposed to be a beacon of freedom in the dark; the angels of liberty who would pursue the men who defile this world and oppress the weak, so that all the world would know, that no matter how small you are, or how weak, or scared; shivering, cold, and alone...that the Americans will come for you. That there is someone out there who will not abandon you. That you are a human being and you deserve more than these endlessly repeated failures of our species.

A killer AI has gone on a rampage through Pakistan, slaughtering perhaps thousands of people. At least that’s the impression you’d get if you read this report from Ars Technica (based on NSA documents leaked by The Intercept), which claims that a machine learning algorithm guiding U.S. drones – unfortunately named ‘SKYNET’ – could have wrongly targeted numerous innocent civilians.

Let’s start with the facts. For the last decade or so, the United States has used unmanned drones to attack militants in Pakistan. The number of kills is unknown, but estimates start at over a thousand and range up to maybe four thousand. A key problem for the intelligence services is finding the right people to kill, since the militants are mixed in with the general population and not just sitting in camp together waiting to be bombed...

 

[VivaRevolution]

It doesn't tell us what human intelligence is used, or how the AI data is fused with human intelligence, but the fact that we don't get to ever know any of that, seems like a fucking problem.

 

[dontsnitch]

The convenient thing about using unmanned drones to kill people is that you can blame the drones when you are criticized. It was the "Algorithm".

 

[JDragon]

Autonomous killer robots violate Asimov's first Law of Robotics. Just saying.

 

[INTERL0PER]

Yup. That's some depressing shit.

When I read an article like that, I often think about the incredible contrast in human lives in 2016. Here I am, living on an island in the south pacific, comfortable and safe... Meanwhile, people in Pakistan are being killed by flying robots. It's surreal and absurd. It's such a reminder to me of what a great life I have, and how all the little bullshit things that annoy me day-to-day do not matter in the slightest. I can just think to myself, "At least I don't have to deal with murderous flying robots gunning me down."

 

[Madmick]

The convenient thing about using unmanned drones to kill people is that you can blame the drones when you are criticized. It was the "Algorithm".

Couldn't even be bothered to read the article. Too chock full of wisdom for that, I suppose.

 

[dontsnitch]

Couldn't even be bothered to read the article. Too chock full of wisdom for that, I suppose.

I don't post in a thread if I don't read at least one source just to avoid unnecessary ridicule.

Anyway, this is a common tactic of deflection. Take the focus away from the killing of thousands of people by pointing to percentages and algorithms. They do it 'cause it works.

 

[Selukrid612]

My position on these kinds of things is this: I'm cool with whatever crazy shit the government wants to do, as long as we get a good movie about it down the road.

 

[MicroBrew]

I read this a few weeks ago and I think it is leaving out a huge factor, namely the Human Intelligence that is also used to identify suspects.

The articles on this program states that the Feds used the example of an Al Jazzera reporter to highlight the efficacy of their software and the critics contend that example proves the problem with this program. I don't think the critics are right because the program squirreled out a person who was in contact with the Taleban and or Al Qaeda, and the person didn't actually get marked for assasination because the analysts could see he was a reporter.

 

[VivaRevolution]

I read this a few weeks ago and I think it is leaving out a huge factor, namely the Human Intelligence that is also used to identify suspects.

The articles on this program states that the Feds used the example of an Al Jazzera reporter to highlight the efficacy of their software and the critics contend that example proves the problem with this program. I don't think the critics are right because the program squirreled out a person who was in contact with the Taleban and or Al Qaeda, and the person didn't actually get marked for assasination because the analysts could see he was a reporter.

But they did create a slide calling the guy a confirmed AQ member, when no info has been offered to confirm this.

 

[Madmick]

I don't post in a thread if I don't read at least one source just to avoid unnecessary ridicule.

Anyway, this is a common tactic of deflection. Take the focus away from the killing of thousands of people by pointing to percentages and algorithms. They do it 'cause it works.

See, this is the part where I know you didn't read either article.

 

[VivaRevolution]

My position on these kinds of things is this: I'm cool with whatever crazy shit the government wants to do, as long as we get a good movie about it down the road.

If not trolling, and there are many more of you, I think you are going to get the exact future you deserve.

 

[Ruprecht]

So did Ars Technica respond to the Guardian article?
Pretty fucked up sense of humour to call it Skynet. Asking for trouble (politically I mean, not in the sense of an AI uprising).

 

[dontsnitch]

See, this is the part where I know you didn't read either article.

Haha. That's not the only conclusion. I could just be slow.

 

[Madmick]

So did Ars Technica respond to the Guardian article?
Pretty fucked up sense of humour to call it Skynet. Asking for trouble (politically I mean, not in the sense of an AI uprising).

Don't believe so. Just ran a few Google/AT searches with date framing and all the keywords. Nothing newer.

Which would appear to validate the Guardian's corrective analysis.

Haha. That's not the only conclusion. I could just be slow.

I'll spell it out for you. The US government didn't "point" to this data. They in no way shape or form desired this data to ever see the light of day, or for these deductions to come together.

Please maybe read the two paragraphs I quoted for my lead-in to the Guardian article. You didn't even have to click the link to know that.

 

[MicroBrew]

But they did create a slide calling the guy a confirmed AQ member, when no info has been offered to confirm this.

Didn't see that mention in the article I read yesterday. They didn't kill him though, or the article would have mentioned it.

I think it is exceedingly unlikely that the US isn't also relying on human intel and other means of electronic surveillance. This program is just 1 part of the vast toolkit the government has.

 

[dontsnitch]

Don't believe so. Just ran a few Google/AT searches with date framing and all the keywords. Nothing newer.

Which would appear to validate the Guardian's corrective analysis.

I'll spell it out for you. The US government didn't "point" to this data. They in no way shape or form desired this data to ever see the light of day, or for these deductions to come together.

Please maybe read the two paragraphs I quoted for my lead-in to the Guardian article. You didn't even have to click the link to know that.

I thought this was how the US responds to criticisms, no? They call them precise by citing percentages- 0.008%.

How else can they condone the killing innocent people?

 

[ShinkanPo]

A thousand? That is a genocide.

 

[ShinkanPo]

@sadclown will love this thread.

 

[Fawlty]

This is horrible and might be enough for a civilized society to purge its leaders. My guess is that it will barely even be a controversy.