First ever AI Drone Swarm used in battle.

[Pauls' Revere]

https://www.the-sun.com/news/3222879/israel-first-ever-drone-swarm-hamas/

SEEK AND DESTROY Israel uses first-ever AI drone swarm in battle to hunt down and blitz Hamas terrorists with NO human input
Tariq Tahir
12:47 ET, Jul 5 2021

Hamas began firing rockets into Israel after protests by Palestinians in May, prompting an 11 day conflict in which 256 people were killed in Gaza and 13 in Israel.

During the violence the Israeli military says that more than 4,300 rockets were fired from Gaza towards towns and cities.

Israel retaliated with air strikes and artillery but didn’t deploy ground forces in its battle with the terrorists.

It now emerged Israeli forces used drone swarms in to target Hamas, New Scientist reports.

Arthur Holland of the United Nations Institute for Disarmament Research said that “if confirmed, they are certainly a notch up in the incremental growth of autonomy and machine-to-machine collaboration in warfare".

Drones have previously been directed by a single operator who ‘fly’ the aircraft from a remote base.

But in recent years, militaries have been working on developing Artificial Intelligence that allows the drones to work together without the need for an operator.

 

[navy-vet]

https://www.the-sun.com/news/3222879/israel-first-ever-drone-swarm-hamas/

SEEK AND DESTROY Israel uses first-ever AI drone swarm in battle to hunt down and blitz Hamas terrorists with NO human input
Tariq Tahir
12:47 ET, Jul 5 2021


Hamas began firing rockets into Israel after protests by Palestinians in May, prompting an 11 day conflict in which 256 people were killed in Gaza and 13 in Israel.

During the violence the Israeli military says that more than 4,300 rockets were fired from Gaza towards towns and cities.

Israel retaliated with air strikes and artillery but didn’t deploy ground forces in its battle with the terrorists.

It now emerged Israeli forces used drone swarms in to target Hamas, New Scientist reports.

Arthur Holland of the United Nations Institute for Disarmament Research said that “if confirmed, they are certainly a notch up in the incremental growth of autonomy and machine-to-machine collaboration in warfare".

Drones have previously been directed by a single operator who ‘fly’ the aircraft from a remote base.

But in recent years, militaries have been working on developing Artificial Intelligence that allows the drones to work together without the need for an operator.

wow, now the drones are seeking out targets autonomusly! How I wonder are they doing this? Facial recognition of known cataloged terrorists, selecting targets that are armed, or just randomly going in and taking out any humanoid?

 

[Brian4Liberty]

wow, now the drones are seeking out targets autonomusly! How I wonder are they doing this? Facial recognition of known cataloged terrorists, selecting targets that are armed, or just randomly going in and taking out any humanoid?

If past history is any guide, the way it works will be with a derived confidence code of some kind. For example, on a scale of 1-10, drones can be set to not fire unless they get a 9 confidence code (enemy or not). The problem being that the people who set these levels will have no concept of what that number means or how it is derived. Often, levels as high as 8 will actually be nearly meaningless. A roll of the dice. But those who issue the orders will be clueless.

 

[Pauls' Revere]

wow, now the drones are seeking out targets autonomusly! How I wonder are they doing this? Facial recognition of known cataloged terrorists, selecting targets that are armed, or just randomly going in and taking out any humanoid?

probably a mixture of both but I would certainly say facial recognition. Some drones, rather the tech they use can even accomplish facial recognition with a mask on.

As worrisome as this is, it wont be long before this tech is used for policing around the globe. Some country, somewhere is going to get a hold of this (if they haven't already) and deploy it against their political opponents. Who, as you eluded to will already be databased in a facial recognition database. I can think of one such country and place off the top of my head. China and the Uyghers and possibly even Hong Kong.

This kind of shit makes me seem like I'm living in the beginning of the Terminator Wars and trying to give people the heads up just simply puts me in the crazy camp.

https://www.zdnet.com/article/facial-recognition-now-algorithms-can-see-through-face-masks/

Facial recognition: Now algorithms can see through face masks
The US Department of Homeland Security has carried out trials to test whether facial recognition algorithms could correctly identify masked individuals.

 

[Pauls' Revere]

If past history is any guide, the way it works will be with a derived confidence code of some kind. For example, on a scale of 1-10, drones can be set to not fire unless they get a 9 confidence code (enemy or not). The problem being that the people who set these levels will have no concept of what that number means or how it is derived. Often, levels as high as 8 will actually be nearly meaningless. A roll of the dice. But those who issue the orders will be clueless.

And I just thought of this.

Let's add social media and social scoring to the mix. We know China uses social scoring based on a number of factors such as what you buy (facial recognition payment systems, Face Pay) and more importantly who you associate with can even increase or decrease your score. So, if your after a know terrorist or someone labelled a such punch in their data and you would have all their associates as well and locales where they have been. Through process of elimination you could (rather the drone) would narrow down the possibilities of where you are thus increasing the success rate of locating the target.

https://www.nngroup.com/articles/face-recognition-pay/

Making Cutting-Edge Technology Approachable: A Case Study of Facial-Recognition Payment in China
Summary: Facial-recognition payment is convenient, but users were concerned and confused after using it for the first time. Better onboarding experiences can help relieve concerns and form factual mental models.

Recently, facial-recognition payment (FRP, or Scan the face to pay, 刷脸支付) has gained popularity in China as a new digital-payment method at physical stores.

This is how they can database everyone. Want to buy food? Show your face to pay. Want the new iphone? show your face to pay. Ugh ...

 

[navy-vet]

probably a mixture of both but I would certainly say facial recognition. Some drones, rather the tech they use can even accomplish facial recognition with a mask on.

As worrisome as this is, it wont be long before this tech is used for policing around the globe. Some country, somewhere is going to get a hold of this (if they haven't already) and deploy it against their political opponents. Who, as you eluded to will already be databased in a facial recognition database. I can think of one such country and place off the top of my head. China and the Uyghers and possibly even Hong Kong.

This kind of $#@! makes me seem like I'm living in the beginning of the Terminator Wars and trying to give people the heads up just simply puts me in the crazy camp.

https://www.zdnet.com/article/facial-recognition-now-algorithms-can-see-through-face-masks/

Facial recognition: Now algorithms can see through face masks
The US Department of Homeland Security has carried out trials to test whether facial recognition algorithms could correctly identify masked individuals.[/QUOTE

Yah, when I first saw this I thought of the Terminator too...

 

[ClaytonB]

If past history is any guide, the way it works will be with a derived confidence code of some kind. For example, on a scale of 1-10, drones can be set to not fire unless they get a 9 confidence code (enemy or not). The problem being that the people who set these levels will have no concept of what that number means or how it is derived. Often, levels as high as 8 will actually be nearly meaningless. A roll of the dice. But those who issue the orders will be clueless.

The training data is probably a bunch of images of things like the Pope labeled as "NOT A TARGET" and then a bunch of images of brown kids looking shifty labeled "TARGET". Turban = TARGET. 1990's Toyota pickup = TARGET. Goatherder = FLIP A COIN. Run the training data through the neural net and voila...

 

[ClaytonB]

probably a mixture of both but I would certainly say facial recognition. Some drones, rather the tech they use can even accomplish facial recognition with a mask on.

As worrisome as this is, it wont be long before this tech is used for policing around the globe. Some country, somewhere is going to get a hold of this (if they haven't already) and deploy it against their political opponents. Who, as you eluded to will already be databased in a facial recognition database. I can think of one such country and place off the top of my head. China and the Uyghers and possibly even Hong Kong.

This kind of $#@! makes me seem like I'm living in the beginning of the Terminator Wars and trying to give people the heads up just simply puts me in the crazy camp.

https://www.zdnet.com/article/facial-recognition-now-algorithms-can-see-through-face-masks/

Facial recognition: Now algorithms can see through face masks
The US Department of Homeland Security has carried out trials to test whether facial recognition algorithms could correctly identify masked individuals.

Meh, the article's title is clickbait. 77% accuracy is garbage and that's one of the better ID rates they were able to get. It really depends on a lot of factors. Gait identification is much more useful against masked individuals. Facial-recognition proper, however, is just a waste of time.

The Chinese are learning this the hard way... A model who has appeared on bus advertisements received countless infractions for jaywalking after the facial-recognition cameras kept firing off every time her face was seen in a crosswalk on the side of a bus. Humorous but illustrates that humanity has an enormous learning-curve to ascend... let's just call it a cliff and we're flying into it at 600MPH...

 

[Pauls' Revere]

Meh, the article's title is clickbait. 77% accuracy is garbage and that's one of the better ID rates they were able to get. It really depends on a lot of factors. Gait identification is much more useful against masked individuals. Facial-recognition proper, however, is just a waste of time.

The Chinese are learning this the hard way... A model who has appeared on bus advertisements received countless infractions for jaywalking after the facial-recognition cameras kept firing off every time her face was seen in a crosswalk on the side of a bus. Humorous but illustrates that humanity has an enormous learning-curve to ascend... let's just call it a cliff and we're flying into it at 600MPH...

LOL that is hilarious. Like strapping my cell phone to a stray dog and see if the wife can find me.

 

[tebowlives]

Can't someone just use an electromagnetic pulse?

Or just put on a Minnesota Vikings uniform since no one who ever wore it could bring harm to even a flea.

 

[Invisible Man]

If past history is any guide, the way it works will be with a derived confidence code of some kind. For example, on a scale of 1-10, drones can be set to not fire unless they get a 9 confidence code (enemy or not). The problem being that the people who set these levels will have no concept of what that number means or how it is derived. Often, levels as high as 8 will actually be nearly meaningless. A roll of the dice. But those who issue the orders will be clueless.

Cluelessness may be part of it, but lack of concern for collateral damage is the main factor. I bet that even with a low confidence roll of the dice for identifying an enemy, they still kill no more indiscriminately (and probably less indiscriminately) than a large number of other alternatives that militaries of the world have long accepted. Anyone who could stomach aerial bombing in civilian areas can stomach this no matter how inaccurate it is.