4 ms·

AI is not allowed to make the decision to kill, that must be done by a human. Having every other step done by an AI is still going to be a huge advantage over n

AI is not allowed to make the decision to kill, that must be done by a human. Having every other step done by an AI is still going to be a huge advantage over non-AI military forces.

michaelmior 3y ago
Even if the decision is made by a human, it really matters how that decision is presented. My understanding of the current state of warfare (which could be way off) is that were possible, targets are still largely selected manually based on verified intelligence. What if we allow AI to propose targets? Even if testing shows that the AI is highly effective at picking "good" targets, that changes the situation entirely. Suppose humans make the final decision of which target to strike. How much time is given to make that decision? What information is presented to them? If the answer is that humans make a decision in a short amount of time largely relying on an AI-generated confidence score, are humans really making the decision?
- Keegs 3y ago
  Exactly. It’s too easy for these decisions to become a formality. ProPublica’s reporting on Cigna[1] is a great example: > Before health insurers reject claims for medical reasons, company doctors must review them, according to insurance laws > A Cigna algorithm flags mismatches between diagnoses and what the company considers acceptable tests and procedures for those ailments Sounds fine, except > “We literally click and submit,” one former Cigna doctor said. “It takes all of 10 seconds to do 50 at a time.” [1]: https://www.propublica.org/article/cigna-pxdx-medical-health-insurance-rejection-claims https://www.propublica.org/article/cigna-pxdx-medical-health...
- hammyhavoc 3y ago
  Have you heard about https://www.theguardian.com/technology/2022/nov/20/part-of-the-kill-chain-how-can-we-control-weaponised-robots https://www.theguardian.com/technology/2022/nov/20/part-of-t... ? And also, how do you feel about already-existing heatseekers? A little different, but the killchain still leans on an automated system for targeting and engagement.
  - michaelmior 3y ago
    Thanks for the link! As far as existing heat-seekers, I would assume that they are generally used in cases where the only target which will likely be locked on to is the desired target. However, this is coming from a place of total ignorance of what the process of target acquisition might look like and what conditions they are used in.
    - hammyhavoc 3y ago
      Heatseekers can be very effective, but as with any automated system, it isn't infallible. Friendly-fire even happens by human beings using small arms, just like strikes on civilian schools and hospitals happen because the intelligence told a human being that it looked like it could be a "terrorist training ground". I'd be interested to see what the failure rate of an AI looked like in what actions it would have taken based on available data versus what actions human took over a several year sample. I have a feeling that the AI will either look terrible, or the human beings will look terrible, or they'll look pretty equal with strange fringe cases where the AI is better than the human, and vice versa. Judgement and authorization are interesting. Something else you might be interested in (which you may already know about) is PAL systems for nuclear weapons: https://en.wikipedia.org/wiki/Permissive_action_link https://en.wikipedia.org/wiki/Permissive_action_link You'll likely be interested in the "two-man rule": https://en.wikipedia.org/wiki/Two-man_rule https://en.wikipedia.org/wiki/Two-man_rule
      - michaelmior 3y ago
        I guess my point was given my limited knowledge, it doesn't seem as though heat-seekers are necessarily any less fallible than humans. I'm not suggesting that "no worse than a human" should be the goal, but I'd say that's the bare minimum.
        hammyhavoc 3y ago
        Precisely. All of these things can fail, with or without human involvement, and humans can fail just as easily. Whilst these are all absolutely horrible contraptions that shouldn't be necessary in relative modernity, it's important to look at stats, but also to "sanity-check" an authorization with concepts like a two-man rule. Whilst AI may indeed be superior to human beings in x areas now or in the future, human ethics, intuition et al are also very important and likely to never be replaced. In the same breath, fuckups will always happen by the very nature of every system and every human being imperfect.
- throw_nbvc1234 3y ago
  identify a target decide to attack the target implement the attack determine the result of the attack If humans are required for the decision, then humans have an inherent hook to prevent more attacks. Otherwise you run into scenarios like https://www.youtube.com/watch?v=pyMNIFZTQkg https://www.youtube.com/watch?v=pyMNIFZTQkg where humans are incapacitated and AI just continue on like nothing has happened. It makes the failure modes more sane. It's not a catch-all but a good agreeable step to take.
Dma54rhs 3y ago
Mines have always made the decision, dumb "AI" based on weight mainly. People do decide where to put them though.
- alwaysbeconsing 3y ago
  Yes, and they tend to cause a lot of harm to non-combatants.
r00fus 3y ago
When every other part of a military decision chain involves AI, the military that wins is the one who can learn to poison or subvert enemy AI (prompt hijacking, etc). The race is on. Both to refine an AI as well as learn to confound them.
DrBenCarson 3y ago
How do you define "decision?" If I tell the AI "Go kill Joe," is that enough? What if I say to "Go kill the enemy." Do I have to visually confirm? Click a button? Locate Joe myself?

AI is not allowed to make the decision to kill, that must be done by a human. Having every other step done by an AI is still going to be a huge advantage over non-AI military forces.

michaelmior 3y ago

Even if the decision is made by a human, it really matters how that decision is presented. My understanding of the current state of warfare (which could be way off) is that were possible, targets are still largely selected manually based on verified intelligence. What if we allow AI to propose targets? Even if testing shows that the AI is highly effective at picking "good" targets, that changes the situation entirely. Suppose humans make the final decision of which target to strike. How much time is given to make that decision? What information is presented to them? If the answer is that humans make a decision in a short amount of time largely relying on an AI-generated confidence score, are humans really making the decision?

Keegs 3y ago
Exactly. It’s too easy for these decisions to become a formality. ProPublica’s reporting on Cigna[1] is a great example: > Before health insurers reject claims for medical reasons, company doctors must review them, according to insurance laws > A Cigna algorithm flags mismatches between diagnoses and what the company considers acceptable tests and procedures for those ailments Sounds fine, except > “We literally click and submit,” one former Cigna doctor said. “It takes all of 10 seconds to do 50 at a time.” [1]: https://www.propublica.org/article/cigna-pxdx-medical-health-insurance-rejection-claims https://www.propublica.org/article/cigna-pxdx-medical-health...
hammyhavoc 3y ago
Have you heard about https://www.theguardian.com/technology/2022/nov/20/part-of-the-kill-chain-how-can-we-control-weaponised-robots https://www.theguardian.com/technology/2022/nov/20/part-of-t... ? And also, how do you feel about already-existing heatseekers? A little different, but the killchain still leans on an automated system for targeting and engagement.
- michaelmior 3y ago
  Thanks for the link! As far as existing heat-seekers, I would assume that they are generally used in cases where the only target which will likely be locked on to is the desired target. However, this is coming from a place of total ignorance of what the process of target acquisition might look like and what conditions they are used in.
  - hammyhavoc 3y ago
    Heatseekers can be very effective, but as with any automated system, it isn't infallible. Friendly-fire even happens by human beings using small arms, just like strikes on civilian schools and hospitals happen because the intelligence told a human being that it looked like it could be a "terrorist training ground". I'd be interested to see what the failure rate of an AI looked like in what actions it would have taken based on available data versus what actions human took over a several year sample. I have a feeling that the AI will either look terrible, or the human beings will look terrible, or they'll look pretty equal with strange fringe cases where the AI is better than the human, and vice versa. Judgement and authorization are interesting. Something else you might be interested in (which you may already know about) is PAL systems for nuclear weapons: https://en.wikipedia.org/wiki/Permissive_action_link https://en.wikipedia.org/wiki/Permissive_action_link You'll likely be interested in the "two-man rule": https://en.wikipedia.org/wiki/Two-man_rule https://en.wikipedia.org/wiki/Two-man_rule
    - michaelmior 3y ago
      I guess my point was given my limited knowledge, it doesn't seem as though heat-seekers are necessarily any less fallible than humans. I'm not suggesting that "no worse than a human" should be the goal, but I'd say that's the bare minimum.
      - hammyhavoc 3y ago
        Precisely. All of these things can fail, with or without human involvement, and humans can fail just as easily. Whilst these are all absolutely horrible contraptions that shouldn't be necessary in relative modernity, it's important to look at stats, but also to "sanity-check" an authorization with concepts like a two-man rule. Whilst AI may indeed be superior to human beings in x areas now or in the future, human ethics, intuition et al are also very important and likely to never be replaced. In the same breath, fuckups will always happen by the very nature of every system and every human being imperfect.
throw_nbvc1234 3y ago
identify a target decide to attack the target implement the attack determine the result of the attack If humans are required for the decision, then humans have an inherent hook to prevent more attacks. Otherwise you run into scenarios like https://www.youtube.com/watch?v=pyMNIFZTQkg https://www.youtube.com/watch?v=pyMNIFZTQkg where humans are incapacitated and AI just continue on like nothing has happened. It makes the failure modes more sane. It's not a catch-all but a good agreeable step to take.

Dma54rhs 3y ago

Mines have always made the decision, dumb "AI" based on weight mainly. People do decide where to put them though.

alwaysbeconsing 3y ago
Yes, and they tend to cause a lot of harm to non-combatants.

r00fus 3y ago

When every other part of a military decision chain involves AI, the military that wins is the one who can learn to poison or subvert enemy AI (prompt hijacking, etc). The race is on. Both to refine an AI as well as learn to confound them.

DrBenCarson 3y ago

How do you define "decision?" If I tell the AI "Go kill Joe," is that enough? What if I say to "Go kill the enemy." Do I have to visually confirm? Click a button? Locate Joe myself?