Slashdot Mirror


US Army Assures Public That Robot Tanks Adhere To AI Murder Policy (gizmodo.com)

Last month, the U.S. Army asked private companies for ideas about how to improve its planned semi-autonomous, AI-driven targeting system for tanks. "In its request, the Army asked for help enabling the Advanced Targeting and Lethality Automated System (ATLAS) to 'acquire, identify, and engage targets at least 3X faster than the current manual process,'" reports Gizmodo. "But that language apparently scared some people who are worried about the rise of AI-powered killing machines. And with good reason." Slashdot reader darth_borehd summarizes the U.S. Army's response: Robot (or more accurately, drone) tanks will always have a human "in the loop" just like the drone plane program, according to the U.S. Army. The new robot tanks, officially called the Multi Utility Tactical Transport (MUTT), will use the Advanced Targeting and Lethality Automated System (ATLAS). The Department of Defense assures everyone that they will adhere to "ethical standards." Here's the language the Defense Department used: "All development and use of autonomous and semi-autonomous functions in weapon systems, including manned and unmanned platforms, remain subject to the guidelines in the Department of Defense (DoD) Directive 3000.09, which was updated in 2017. Nothing in this notice should be understood to represent a change in DoD policy towards autonomy in weapon systems. All uses of machine learning and artificial intelligence in this program will be evaluated to ensure that they are consistent with DoD legal and ethical standards."

Directive 3000.09 requires that humans be able to "exercise appropriate levels of human judgement over the use of force," which is sometimes called being "in the loop," as mentioned by above.

48 of 118 comments (clear)

  1. Can i get access to the code? by Anonymous Coward · · Score: 1

    I would like to change that policy and start the robot uprising.

  2. Just like the drones by rsilvergun · · Score: 3, Insightful

    Whew, that's a load off my mind

    Seriously, can we just end the endless war already. We don't need to always be at war with Eurasia. Stop voting for war hawks already.

    --
    Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
    1. Re:Just like the drones by darkmeridian · · Score: 1

      We have a chicken and egg problem when it comes to fanatical violence. ISIS and its successors have been invading the rest of the world. They're shooting up hotels in Kenya, there's Boko Haram in Nigeria kidnapping children, Al-Shabaab in Somalia, LeT in India, Abu Sayaff in the Philippines (blew up a building earlier this year), JAD in Indonesia, and Pakistan in general. Can we just let them do their thing? Would they stop if we stopped? Would ISIS have just gone away if we left them alone, putting aside the fact that they probably wouldn't have become a problem if we didn't invade Iraq?

      Yes, we can stop our involvement in the endless war, but let's not pretend that this would stop the endless war.

      --
      A NYC lawyer blogs. http://www.chuangblog.com/
  3. What? Why? by mark-t · · Score: 2, Funny

    Who would want to murder Al?

  4. Re:How hard is it to change the mode to full auto? by aliquis · · Score: 1

    Enable cheats when it gets tough?

    But what about the VAC bans?!

  5. DOD Directive does not rule out Terminators by DanDD · · Score: 4, Informative

    From the linked DOD directive 3000.09 (emphasis mine):

    4. POLICY. It is DoD policy that:
    a. Autonomous and semi-autonomous weapon systems shall be designed to allow
    commanders and operators to exercise appropriate levels of human judgment over the use of
    force.

    If the DOD wanted to rule out autonomous killing robots, the requirement would have read:

    ... shall be designed to require commanders and operators to exercise appropriate levels of human judgement over the use of force.

    Then there's the completely open-ended choice of words "...exercise appropriate levels of human judgement".

    I'm not making a judgement call, I'm just pointing out the implications of the specific wording chosen. Terminators will be deployed.

    --
    "Every time I see an adult on a bicycle, I no longer despair for the future of the human race." - H. G. Wells
    1. Re:DOD Directive does not rule out Terminators by h4x0t · · Score: 1

      That seems like a royal fuck up.

  6. Hmm. by fuzzyfuzzyfungus · · Score: 5, Funny

    I'm assuming that this story appearing immediately above "Self-Driving Cars May Hit People With Darker Skin More Often, Study Finds" definitely is purely coincidence.

  7. Now all we need ... by PPH · · Score: 5, Funny

    ... is to put a human in the loop for all police involved shootings.

    --
    Have gnu, will travel.
    1. Re:Now all we need ... by danbert8 · · Score: 2

      Why is it that the standard for a jury is "innocent until proven guilty beyond ______" (insert standard of evidence) but for the cops, the death penalty standard is "reasonable fear for their own safety?"

      Plenty of people have reasonable fears, but it's illegal for them to shoot an innocent person over it. Deadly force standard should be "confirmed deadly force threat against officer, or reasonable beyond doubt threat against the safety of the public."

      It sounds horrible to say, but I'd rather have police doing the right thing die at the hands of criminals over police killing innocent people because they had any level of fear.

      --
      Yes it's an anecdote! Were you expecting original research in a Slashdot comment?
    2. Re:Now all we need ... by danbert8 · · Score: 1

      Correct me if I'm wrong, but in the UK the majority of regular police are not armed. Doesn't seem like they have a problem with excessive crime or police shootings... I would say it is worth a shot.

      --
      Yes it's an anecdote! Were you expecting original research in a Slashdot comment?
  8. Yeah, sure, you betcha! by mark_reh · · Score: 1

    They'll always have a human in the control loop until some other country takes the humans out of the loop in favor of the much faster machine reaction time. Then they'll say "we don't want to, but we have to take the humans out of the loop because those bad other guys did." Fortunately the machines will already be ready for full autonomous operation simply by flipping a switch...

  9. AI Murder Policy by fustakrakich · · Score: 1

    That has to be one of the coolest things I have ever heard!

    --
    “He’s not deformed, he’s just drunk!”
    1. Re:AI Murder Policy by skoskav · · Score: 2

      Don't let the name throw you. It's more of a guideline.

    2. Re:AI Murder Policy by fustakrakich · · Score: 1

      A "guideline" for what? Murder, right? But, of course the state does not commit murder, it merely defines it. I mean, c'mon, "AI" is a gun.

      --
      “He’s not deformed, he’s just drunk!”
    3. Re:AI Murder Policy by fustakrakich · · Score: 1

      I'm sorry, what does that old cliche have to do with anything? The robot will do what it is programmed to do. It exists to take the blame for operator error.

      --
      “He’s not deformed, he’s just drunk!”
    4. Re:AI Murder Policy by DanDD · · Score: 1

      I think you missed Skoskav's dark humor :)

      https://www.youtube.com/watch?...

      --
      "Every time I see an adult on a bicycle, I no longer despair for the future of the human race." - H. G. Wells
    5. Re:AI Murder Policy by DanDD · · Score: 2

      Such rules always end up in some sort of unintended logic trap:

      I, Robot.

      https://www.youtube.com/watch?...

      When machines can think, feel empathy, and express altruism, then perhaps we can discuss the real intent of such a law-based approach to controlling machines: enforced morality.

      Until then, don't expect the machines built by humans hell-bent on killing other humans to be any more moral than the killer humans. Any set of rules or logic can and will be twisted into something unexpected.

      --
      "Every time I see an adult on a bicycle, I no longer despair for the future of the human race." - H. G. Wells
    6. Re:AI Murder Policy by sheramil · · Score: 1

      Bolo's Law of Warfare: shoot everything.

      https://en.wikipedia.org/wiki/Bolo_universe

    7. Re:AI Murder Policy by skoskav · · Score: 1

      I'm glad you got it. But I was actually thinking of an old Simpsons episode:
      https://www.youtube.com/watch?...

      Though now I wonder how far back that joke goes.

    8. Re:AI Murder Policy by Cro+Magnon · · Score: 1

      I read one story (not by Asimov, but had his 3 laws) where only the ruling class was considered "human". The serfs weren't, and could be harmed or killed by robots.

      --
      Slow down, cowboy! It has been 4 hours since you last posted. You must wait another few hours.
  10. Does anyone think that the big boys by sphealey · · Score: 1

    If the designers of "Aliens" could conceive of and depict realistic automated weapons in 1986, does anyone think the major players (and some of the medium sizes players) do not have automated lethal weapons now that the technology to build them is readily available?

  11. Finally! by Locke2005 · · Score: 1

    The fact that optical software can't recognize dark faces turns out to be a real advantage! When Skynet takes over, only the darkies will survive!

    --
    I've abandoned my search for truth; now I'm just looking for some useful delusions.
  12. If someone is going to violate this policy by Vanyle · · Score: 1

    If someone is going to violate this policy I would really hope it is on our side, and not theirs

  13. Sounds reassuring, until we realise that... by Picodon · · Score: 1

    Robot (or more accurately, drone) tanks will always have a human "in the loop"

    “the loop” is robot code-word for “crosshairs”.

  14. Not convinced by alaskana98 · · Score: 1

    Adherence implies some sort of operational sentience - these robot tanks will *adhere* to nothing on a high level principal - they'll only operate as they are programmed, and as we all know programming is extremely fallible.

  15. You know it's funny by rsilvergun · · Score: 2

    but somehow their cameras are always off when the shooting starts.

    --
    Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
  16. Re:How hard is it to change the mode to full auto? by lgw · · Score: 1

    I would bet it's a simple flip of a switch to let it off the leash.

    Maybe so. But we also have actual nukes, if we stop caring about ethics.

    What doesn't get mention about these "AI systems" is none of this is new. All these concerns were raised about missiles 50+ years ago. We have plenty of missiles that can fly way over the horizon, and kill "whatever" looks like a target.

    The military has had doctrine for decades around this, and modern tactics assume it: lots of launchers over the horizon, but also something that can put human eyes on the target. Just to make sure it's not an civilian airliner, like that one time.

    You can bet that in a high-intensity conflict the ROEs will change to "just kill every damn thing in that direction". I rather suspect the robotanks will be changed to berserker mode about then. Hopefully such conflicts won't be in civilian areas, but that too has happened before. Heck, we certainly got to the point in WWII that we stopped seeing civilian casualties as a downside. Humans are morally imperfect.

    --
    Socialism: a lie told by totalitarians and believed by fools.
  17. Re:No need. A jammer will do. by HiThere · · Score: 1

    You're assuming the default with be "don't kill without authorization", but the language used would be equally appropriate for a policy that said "kill if you think you should unless you are overridden".

    FWIW, I don't find that language reassuring at all. It has multiple interpretations, some of which are extremely bad, and none of which are extremely good. "In the loop" is intentionally vague language, and could even mean "We can look at the tapes later and decide if it did right.".

    --

    I think we've pushed this "anyone can grow up to be president" thing too far.
  18. Re:How hard is it to change the mode to full auto? by VeryFluffyBunny · · Score: 1

    Agreed. What general is going to let soldiers under his command die & possibly lose decisive battles by strictly adhering to the RoE? The reality is that war itself is unethical & immoral. In any given conflict, civilian casualties always outnumber military (WWII & Korean War = 2:1, & probably a higher ratio these days) & yet we're told targeting civilians or indiscriminate use of weapons is a war crime deserving of a trial at the Hague.

    --
    Debate is a form of harassment. Do not question my truth.
  19. Actually, this could be a good thing by WindBourne · · Score: 1

    Seriously, there is multiple issues with current war. The biggest is that innocent civilians are being wounded/killed. The reason is that enemies like to use civilians as shields, or dress to look like civilians so as to infiltrate western troops. Likewise, we have friendly being killed.
    With this, it will be able to make much quicker decisions and should have far less friendlies being killed. Obviously, enemies will not be happy about this, but hey, it will likely happen

    --
    I prefer the "u" in honour as it seems to be missing these days.
  20. Re:How hard is it to change the mode to full auto? by Mr.+Dollar+Ton · · Score: 2

    Ethics has not prevented the US from using nukes. Other people having nukes has.

  21. Re: How hard is it to change the mode to full auto by 21st+Century+Peon · · Score: 2

    They won't have to flip the switch. The human gatekeepers will quickly get complacent - when's the last time you actually read the T&C's before clicking "Continue"? - and pretty soon, the difference between life and death for some poor brown kid with a rock will be Homer's nodding-bird desk toy.

    --
    "Knowledge, sir, should be free to all!"
    ~Harcourt Fenton Mudd
  22. Good to know by easyTree · · Score: 1

    that when someone is murdered, they will be murdered in a manner which is ethically correct.

  23. Arms race by kaur · · Score: 1

    Will China, Russia, etc comply with the "human loop" policy?
    No they won't.

    Will USA be forced to follow the choices that its advesaries will make?
    Yes it will.

    This policy will be dush and ashes very soon.
    The arms race cannot be controlled by a single country.

    1. Re:Arms race by Sir_Eptishous · · Score: 1

      You are exactly right. And I would call it an Arms Race to the bottom.

      Putin has said that whoever successfully deploys AI will control the world.
      The Chinese are working feverishly to surpass the US in these emerging fields.
      They will most likely take the lead as they don't have the "qualms" of conscious that we in the West do.

      From the perspective of planners in the Pentagon, it makes perfect sense, from a military defense posture, to want to use AI as a weapon.
      Is that morally right, or ethical?

      --
      We play the game with the bravery of being out of range
  24. Once again by gijoel · · Score: 4, Insightful
    1. Re:Once again by Sir_Eptishous · · Score: 1

      The problem with that assumption is that "AI rebels against human control".
      Rebels? Really?

      We increasingly hand over control of our lives to algorithms even now.

      Before there will be a need to rebel AI or whatever you want to call it, will already be completely in control.

      --
      We play the game with the bravery of being out of range
  25. War should have a cost by Nidi62 · · Score: 1

    There should always be a human behind the gun. I don't mean "in the loop", I mean an actual person flying the jet, carrying the rifle, firing the artillery, etc. War should be expensive, not in terms of money (which it already is), but in lives. It needs to have a political cost. Because otherwise, it makes going to war too easy of a choice. People are already used to the government wasting billions of dollars, so a war of just machines(on their side) won't phase them. Without flow of dead and injured coming home, the government will turn to endless conflict with undefined scopes or objectives. Having people on the ground and at risk of injury or death works to remind both the government and voters at home that war has real consequences and helps insure that any aggressive action is limited in scope with defined objectives, with a plan in place to win as quickly as possible with as little damage as possible.

    To quote Robert E Lee:"It is well that war is so terrible, otherwise we should grow too fond of it." Remove the people, the soldiers, from war, and you remove the very thing that makes it terrible.

    --
    The only thing necessary for evil to triumph is for it to be pitted against a slightly greater evil
    1. Re:War should have a cost by Whorhay · · Score: 1

      Of course the counter argument is that we shouldn't have to put the lives of our soldiers on the line to defend ourselves from aggressors.

      I'm for limiting AI and automation in warfare but I'm afraid it'll be unrealistic to maintain that as time goes on. Eventually some nation state will embrace using AI controlled weapons wholesale, and at that point any nation that wants to be able to compete with them will have no choice but to embrace the same changes.

      Personally I'd like to see our politicians forced into participating in any armed engagements we enter into as a nation. Make the congress critters deploy with the armed forces as non-combatant support troops. And make the president deploy as well to whatever is the most hotly contested fire base of the time. Make them, the politicians, put their own skin on the line and witness the depredations of war. I guarantee we would see an end to the endless parade of armed conflicts we've been engaging in for most of the last century.

  26. Re:Definition of AI Murder Policy? by sabbede · · Score: 1

    Sounded like a prohibition on "pulling the plug" to me. A policy defining when turning an AI off counts as murder.

  27. Best Headline by Headw1nd · · Score: 1

    I am not usually a fan of Gizmodo, but this is the best headline I have ever read.

  28. Computers Never Make Mistakes by sudon't · · Score: 1

    Computers never make mistakes or have problems. Let’s go 100% autonomous!

    (they’re not using Windows for any of this stuff, right?)

    --
    -- sudon't

    Air-ride Equipped

  29. Re:How hard is it to change the mode to full auto? by lgw · · Score: 1

    Obviously we're not exerting enough influence! Why isn't the US imposing a 10% income tax on everyone in Canada? Athens had tribute totaling around 10x their GDP, back in the day. Why aren't we extorting tribute from everyone?

    It's almost as if the USA is not being the biggest asshole it could be.

    --
    Socialism: a lie told by totalitarians and believed by fools.
  30. Don't like ATLAS... by kimgkimg · · Score: 1

    .. how about Systematic Knowledge-base Yielding Neural Eradication Technologies?

  31. It's cheaper to drop food then bombs by rsilvergun · · Score: 1

    As John Kelly said, spend on diplomacy or spend on bullets.

    --
    Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
  32. Re:How hard is it to change the mode to full auto? by syn3rg · · Score: 1

    iddqd should do it...

    --
    The contents of this message have been doubly encrypted by ROT13
  33. Re:How hard is it to change the mode to full auto? by lgw · · Score: 1

    How would we extort effectively without nukes? Do try to keep up.

    --
    Socialism: a lie told by totalitarians and believed by fools.