Can Morals Be Programmed?

Absolutely not. Autonomous weapons are a bad, BAD idea. To get a sense of what a horrible idea this is watch the (excellent) movie Dr. Strangelove.

Think having your PC slow down because it's infected with a virus is bad? What happens when one of these suckers gets hacked?

The weapons are as good as the information that is inputted into them.

They put mines in the water that only functions with a certain type of ship.

They have torpedoes that follow only a certain type of screw signature.

Land targets have a grid location that is inputted before launch.

A smart missle that is to blow up tanks has the electronics to pick up a certain signature that is given off by that type of tank and ignore the school bus.

Some are outfitted with cameras so the operator can see where the missile is going.

Military has been using "Fire and Forget" Techology for years. There is nothing new going on. You're just hearing more about the remote controlled technology lately then what you're used too.

There will always be a human in the loop for some time to come. Someone will always give the order to attack something or defend something. It's not the machine applying the rules of war, it is the programmer and operator of the machine or more appropriately, the general in charge that is responsible.

However, you will see more and more automation appear as technology advances.

As for the rules of engagement, nothing will change. Each event will be carefully planned and executed. Will mistakes happen? Damn, yes! However, today's execution of warfare (at least from the free world) is orders of magnitude more concerned about civilian or collateral damage than in past wars.

Don't believe it? Look at WWI and WWII. Whole cities with civilian populations were razed, killed, and burned out with non-stop bombing.

However, I don't worry about the free world's morals of war as much as I do about the tyrants' armies that use their own citizens as human shields or deliberately target civilians, babies, women, and children.

Maybe we should stop giving ourselves such an annal exam and start turning our outrage toward the terrorists and tyrants that have no (zero) scruples at all?

Or how about the UN soldiers that saluted the Hezbollah murder of innocent Israeli babies after his release from Israel and subsequent return to Palestine? That one was caught on film (see below), but where is the outrage over the total lack of morals here?

I think we may have a much better chance of programming morals into machines than we will ever have instilling them into some societies and their people.

Not while mankind is doing the programming.

http://cr4.globalspec.com/blogentry/9604/Upcoming-Military-Robot-Could-Feed-on-Dead-Bodies

Morals cannot be defined with 1's and 0's... Being Moral is not being Logical.

However, as long as every "Good" soldier has a transmitter/receiver on their person to identify them on the field as "not a target" it could work in theory.

But all this sounds so familiar... wait, I know, It's right out of the movie "Terminator" (I just watched T1&2 last night).

Humans VS the Machines... Can I get a Digitized Hu-Rah!?

Ah HUUh Ah !!

Gaaak. Rules of war are constantly broken, intentionally or unintentionally. We have "friendly fire" and "collateral damage" coming from human errors/breaches of the "rules of war". We have Abu Ghraib if you want to talk about "programming" morals, what happened there?

Automating vehicles to follow "rules" and make judgement calls where human perception and human morals are known to fail, is a risky business.

At least, with human soldiers, there is the possibility that, once engaged in an error, they will recognize the mistake they have made or the immorality of their own acts and call a halt. There's no such possibility for a machine: I doubt it could be effectively made to re-evaluate the decision to strike during the engagement. It would be all or nothing.

In the end, there will be no human being to face the music for errors made. No one to feel guilt or suffer public blame. A machine feels nothing.

I'm sorry, but most commentators here are saying it can't be done. I assure you it can. and those machines would be very efficient. machines always are. morals can be programmed and can pass the Turing test. I'm sure that the machines can be programmed to be more trustworthy than lots of boy soldiers and gang bangers who already run around with automatic weapons.

The question is, "Can you negotiate with that algorithm?" and what tests will be performed to determine enemy or ally? How can anyone, machine or human distinguish good from evil, enemy from ally, friend or foe. We must know ourselves before we can program that.

The enemy will not always want to identify themselves... and you may not always know who your friends are... but as in all human history, wars are won by deception. Can a robot perceive truth? Can we?

No. I think they should not be programmed to be autonomous. Not because we can't, but because we don't have enough morals and awareness ourselves to do the job properly.

therefore, we should just stick to doing the dirty work ourselves, and continue to pay for our ignorance, if it needs to be done, not amplify it to these terrifying proportions.

Lastly, if the US can do it, what makes you think enemies can't? (ever changing enemies) Would you like to think that robots are programmed to kill you? (just like Terminators)

Every time the US raises the bar, the adversarial countries raise the bar too. this is continuous escalation , whose only purpose is to kill, and thereby defer the eventual consequences of global overpopulation. I think we should put those hundred billion dollars into education, especially of responsible parenthood and negative population growth strategies that don't involve global mass murder, like is being done now. PermaWar is not right.

There are other ways, for anyone with even a glimmer of imagination.

Chris

I agree it will be done/has been done. Here is an autonomous non-lethal defense mechanism in civilian use right here and now.

Can Morals Be Programmed?

They are the program!

Ky.

It may be easier to just program potential targets, and leave the morals to the liveware. Any program or moral can be subverted if there is sufficient pressure. Someone mention Dr Strangelove, but I'd refer to 2001 a space odessy, where HAL's programming was altered to ensure the secrecy of the mission. That was revealed in the book 2010.

The point I'm trying to make is no matter how pure the intention, there is always the opportunity for skulduggery.

I'd rather the decision to end someones life remain with the living.

The point I'm trying to make is no matter how pure the intention, there is always the opportunity for skulduggery.

The program is the morale, like I said.

A machine does what its told to do. With the occasional error from various outside influences like radio interference or poorly manufactured parts.

So how can it have morals. If it had morals it probably wouldn't be killing another human being to begin with.

Its absolutely great "..that soldiers can fire from a 'safe' distance" - However as anyone given thought to the innocent child that is not at a safe distance and that gets blown to pieces.

Governments spend millions to protect those trained to fight but do nothing for the innocent and the weak. What a wonderful world we live in.

Only morals can change programs. Time for a change, I hope!

Like my self, we can twist, but morals have to change and not be implemented. They stay in the blood for too long and changing, although legitimate, involves inertia, and that can take time. That it is our time, is the biggest waste of all. Lets waste less and get down to eradicating indulgence in pussy talk. We are the world, we are the people (sorry Michael).

What we tell a computer is not what he wants to hear. What the computer tells us is what we can listen to, if we have enough time to comprehend.

Morals? Any one? I hope, Ky.

PS: What a wonderful world (sorry Luis)

Awe come on. I am sorry, but that is not only offensive, but utter BS.

If that really outrages you why not start with the mud suckers that use their own women and children as human shields or specifically target women and children or civilians as part of waging their campaigns?

What about those brave "counter insurgents" that force little kids to hold and fire weapons at the advancing infidels, while they flee wearing women's clothes? I've personally talked to the soldiers that were forced to fire back at (and kill) kids that haven't even gone through puberty yet. They can't hold the rifle, so they jam the butt of the rifle into a wall so they can pull the trigger. You have no idea how those soldiers feel about facing that and returning home.

Instead we get an unending drone about collateral damage from soldiers and their commanders that go way, way out of their way to try to prevent that. I am not saying that they don't make mistakes, but there is no comparison about an allied force that goes way beyond the Geneva Conventions as opposed to one that breaks every freaken' rule on the book.

For your information, we do spend millions of dollars to prevent collateral damage and to protect civilian life. I would know, because it is part of my business.

I don't mean to be personal, but your statement is completely wrong and you are not alone in that regard.

What are you on about? If you would actually read what I wrote instead of spewing your anti-everbody BS you might think twice about what you wrote.

I was not specifically targeting your country.

Actually this is typical of the response from you, you jump in with both feet before thinking about what was actually said or intended. You are so afraid that anyone could possibly have a different opinion to you that you immediately start of throwing insults.

As for spending millions to prevent collateral damage - why not ask the mothers and fathers who have picked the bodies of their children up off the street if they are grateful that a couple of million dollars was spent to protect civilian lives. Oh, before you go off on one again, this is not specifically aimed at your country. All the world powers are guilty of this.

"you immediately start of throwing insults."

No insult intended. Please reread my post and read this one.

I wrote, "allied force that goes way beyond the Geneva Conventions as opposed to one that breaks every freaken' rule on the book."

This is not limited to the US. I understand that you were not pinpointing only the US (even though the US does seem to get most of the finger pointing, but that is another subject).

"for spending millions to prevent collateral damage - why not ask the mothers and fathers who have picked the bodies of their children up off the street if they are grateful that a couple of million dollars was spent to protect civilian lives." That's a red herring argument. Nobody wants or likes the idea of people dying, particularly in war, but the truth is that it happens (for a variety of reasons), but to blatantly state or infer that the free world does not spend money and effort (and lots of it) to avert unneeded deaths is absolutely false.

The data will not support your claim, but if you have data that shows otherwise, let's see it. Like I said, I work in the field of training and I personally know for a fact your claim is untrue.

My rant is not against you, it is against ignorance and injustice. I hope you understand what I mean. It is not personal and I stated that earlier and I restate it now; I am not trying to personally attack you.

"why not ask the mothers and fathers who have picked the bodies of their children up off the street if they are grateful that a couple of million dollars was spent to protect civilian lives." That's a red herring argument." - ask the Catholic families of Northern Ireland if this is a red herring.

Yes I agree that these topics often end up in US bashing but that is not my intention. I am just pointing out that the couple of million dollars (according to you) that is spent on protecting civilian lives is quite a small amount when compared to the billions that is spend on a war, in fact I it would just appear just to be a token effort to appease the masses.

I agree.... first appearance is thats less than the cost of socks for the soldiers. (although I don't know if the numbers are real.)

http://www.nytimes.com/2007/01/17/business/17leonhardt.html perhaps this will give an insight into the costs of war. I am sure that other conflicts around the world cost slightly less than the numbers presented in this article but they give rough numbers on the cost of war.

Perhaps the estimated $250Billion veteran health care costs may have something to do with having the soldiers fire from a safe distance. Its not to protect the soldiers, it to reduce the costs, I bet some accountant somewhere worked out the costs and luckily (for the soldiers) it just happenned to be cheaper to invest in this technology rather than have the soldiers killed or maimed. Note there is no such numbers presented for the costs to civilians.

At first a couple of million seems like a huge number and one would think that a government who spends a couple of million on protecting civilian lives are really serious, but when you compare it to the hundreds of billions that are spend on a war then it shows exactly how serious governments are about protecting civilian lives i.e it is no more than a peace meal effort to appease the masses.

Clearly the fact that Anonymous Hero works in this area has affected his judgement somewhat.

the goal of government is to increase government.... that means eventually taking over the world.

now that governments have discovered robotics and remote control, you can expect that governments will continue to invest, and the next wars will be fought by advanced robotics, with very few soldiers on the ground. automated vehicles and robots.

the purpose is to control the assets, resources, and populations of the target country.

there will be RC farms of soldiers sitting in trailers and sea-cans, taking over countries half the world away.

because you are right... real soldiers cost money if they get hurt, and the cost of robotics will continue to decrease, and with moore's law, they will continue to get more powerful and intelligent.

morality? its the golden rule... he who has the gold makes the rules.

Chris

"Clearly the fact that Anonymous Hero works in this area has affected his judgement somewhat."

Absolutely! It is called experience.

1) It's still is a red herring. Actually, it is more like a "Description of Composition". Your claim is that the amount of money spent to protect our soldiers vastly outweighs the money that is spent to protect innocent lives. Actually, you used the words "do nothing."

Your proof is that there are many families who have lost innocent family members that are grieving, therefore the claim, "Governments spend millions to protect those trained to fight but do nothing for the innocent and the weak" must be true.

Obviously, the argument is fallacious. You have established no relationship between grieving and dollars and effort spent by governments. I'm not sure that it can be established, but you have no data to back the claim anyway nor expertise in the field (I am assuming).

However, my field of expertise is military training. It is my business. I have personally designed some of the tools used to do that training and personally participated in various training sessions (many unrelated to my immediate area of business). I know what is in the training programs. I know what the goals and objectives are for those programs. I know what the results of those training programs are. I also have some insight into those program costs.

I may not be the foremost expert in military training, but I can say that I have a very informed opinion on the subject and state that with great confidence because I have the credentials to back that up.

If for no other reason, just trust me on this, your assertion could not be more wrong.

Lets say for 1 minute that you are the expert (self acclaimed I add). Can you, with your expert insight into the workings of government and the military, explain the millions (alleged by you) spent on protecting civilians versus the hundreds of billions spent on war. Work out the percentages and come back with an explanation. I think you will find the percentage very close to zero (very close to zero in my book equals "do nothing")

Since you work in the industry you will not come out and criticize your paymasters and I can understand that (puppet on a string jumps to mind). But do not try to insinuate that the rest of us who have a different opinion are completely stupid.

Lets face it, we could continue this thread for months and we will not agree, so it is best that we wrap this up. If we continue then someone or other will complain to Admin, because something we have said upset their delicate nature/conscience and Admin will stop the thread instead of just ignoring the complaint.

I suppose, in some ways I am as arrogant and self opinionated as you so it is highly unlikely (read impossible) for us to agree.

Take care.

Regards

Mr. W.A Snow

"I think you will find the percentage very close to zero (very close to zero in my book equals "do nothing")"

It's your claim, not mine. So, I'm calling you on it, substantiate your claim.

But are you not the expert in this field (self acclaimed), after all it is your business (stated by you in post 20). Surely you have the figures given your insight into the internal workings of government and the military. It is my opinion that you have the numbers but you are unwilling to share these as it would make your claims a joke.

You were the one that came out with drums beating and boldly stated that governments spend millions (your words, see post 20) on preventing civilian causalities, this knocked us all back and I am sure some thought that, wow Anonymous Hero really has his finger on the pulse and we should believe his rhetoric - he really is an expert and that penguin is a clown.

Now the millions sounded really great until it was pointed out that they also spend hundreds of billions on war (see NY Times article). Now that really makes your millions sound paltry. Even you must realise that. I am sure we can all do the math and see what percentage is spent on saving civilian lives versus the overall war machine.

Or is your argument that the cost of war is not hundreds of billions (as pointed out the the NY Times) but rather a couple of million? Now before you go down that route perhaps it would be worth calculating how many modern war planes you could purchase for a couple of million dollars - lets take a figure of 50 million dollars (now that sounds like a hell of a lot of money, right? just like the millions spend of saving civilian casualties) well for 50million dollars you could probably get 50% of a Eurofighter Typhoon or a Sukhoi FGFA, or a Misubishi F-2. Now the plane on its own (or rather the 50% of the plane in this case) is a bit useless so we now need trained pilots, fuel, transport and last but not least missiles that can be fired from a safe distance to protect the expensive equipment and the highly trained pilot. As you can imagine these things cost money.

I believe that this substantiates my claim that in reality, governments do nothing to protect civilian lives.

The numbers just don't stack up in your favour, do they?

That is not an argument and it does not substantiate your point.

You made the claim! Back it up!

I do not have to substantiate your claim. I just challenged your claim.

Frankly, your claim and subsequent argument is baseless. Anyone with any knowledge on the subject knows that and you are just making yourself look silly. If you had gone out and got an education of the subject first you would have never made the assertion in the first place. Your argument is simply emotion based and you have unfortunately made a mistake on the topic. Why not do some unbiased research on the subject then come back and share what you have learned?

You claim that we should trust you but when asked about the percentage spent on protecting civilians compared against the total spent on a war you have no answer, instead you start to throw insults in the hope that it will divert attention from the issue.

What was it that you said in post 43 "If for no other reason, just trust me on this,.." basically you are asking everyone to trust you and you can't come up with a defence. To be honest it sounds a lot like a used car salesman. I am sure that your paymasters that are pulling the strings are quite happy with you.

Just because someone gets a nice fat pay cheque at the end of the month does not make the slaughter of innocent people right. You really need to understand this.

The numbers have been put forward and you can do nothing better than start throwing insults again (I pointed this out a number of posts back, it is typical of you).

While I have no difficulties in someone being patriotic, I do have an issue when someones patriotism borders on fascism/ racism.

Now when you are willing to debate the subject based on the facts then please come back and perhaps someone else will be willing to engage you. I however, will not as I personally have no time for fascists and racists (this is based on your post 20 which referred to the "mud suckers" - I am sure that this has a derogatory meaning somewhere in the military training manual)

I would like to say that it was a pleasure in debating this subject with you, however when personal insults start being thrown around then there is little room for debate.

I assume English is not your primary language. Could there be a possibility that you do not fully understand what I wrote?

If so, I am sorry that you misunderstood so many things, primarily, the difference between ad hominid attacks (attacking the person) and attacking their argument. It is easy enough for two people speaking the same native languages to be confused on that issue, let alone different languages. I don't know you personally and we could disagree on everything under the Sun, but it would not make you my enemy. I don't know you as a person and will not pass judgement on you personally. I only ask the same in return.

First, my references to mud suckers simply applies to people and governments that endorse the tactics of using their own women and children for human shields and using children as proxies to fight their wars under physical coercion.

So, let's clear the air. Do you feel those actions are justifiable? In other words, do you support that?

Please be specific, where am I being a racist? Where am I being a fascist?

Back to mud suckers. If you feel that it is okay to use civilians as human shield, but feel that the military does not do anything to protect them, then I would think that would be hypocritical. Do you agree?

My personal belief is that governments and people that do use human shield are the lowest form of humanity there is. That is where the term mud suckers come from. Let me be clear. I do not support that, but I would love to hear your opinion on the subject and its relation to racism and fascism. What would you think would be a better term for those people? Is mud suckers too harsh?

I like spirited arguments, too.

I assume English is not your primary language. Could there be a possibility that you do not fully understand what I wrote?

This is intentionally insulting, and weakens your entire argument, because it makes you come across as a typical, uneducated red neck. Snow shows every sign of communicating very well in English. I think it is unlikely that he would, for example, make such an idiotic and ironically funny mistake as your "ad hominid attacks." Are you thinking that war is between primates and weasels?? Or were you suggesting that because "hominem" comes from the Latin language, it is unimportant to use the right word (... any ferrin language being inferior to good ole umercan.)??

Perhaps one good reason for avoiding political or moral discussions on CR4 is that our members are so poorly equipped for such discussions. I assume your engineering is more effective than your use of language or tact.

I appreciate the spirit of your post.

heck, people won't even agree upon which instant " life " is . how can there be a correct answer if such a basic question can't , won't be solved thru rational discussion.

I agree. Calm, rational, respectful discussion can lead to valuable insights and perhaps common understandings. However, that has not been the tone of the dialogue between HoleInTheSnow and AnonymousHero.

Virtually all of us who are not sociopaths can agree that people the world over have a great deal in common, particularly in regard to central moral principals and ethics. Words to the effect of "Thou shalt not kill" are found in religious texts all around the world. Love thy neighbor. Treat others as you would like to be treated. These are universal words of guidance for a life well-lived. Compassion is valued.

Thus, when HoleInTheSnow, in his first post of the thread, expressed concern for "the innocent child" and "the innocent and the weak" he was writing in concert with universal human values. Hero's response, "Awe come on. I am sorry, but that is not only offensive, but utter BS," is combative and insulting, and not the start of a rational discussion. It is the start of a contentious argument... as we have seen.

A discussion (rather than combat) might have started like this: "I can understand your point, HoleInTheSnow, because we share a concern for the lives of innocent victims of warfare. Although there are many examples of military excess, and certainly many thousands of civilian casualties in recent actions by NATO and other forces, I believe that there is a great deal of thought and effort that goes into seeing that collateral damage is minimized. For example... "

imho : w/ o discussions or debate .. isn't that result just dogmatic resignation to the status quo ? the changing of an opinion over any given subject might , that change in perception then alter any further decisions?

I agree. Discussion and reasoned debate can change perceptions, widen awareness, and foster understanding. Verbal combat, on the other hand, typically has the opposite effect, leading to entrenched views.

if science is to explore , examine , then what limits should be placed? by whom ?

By whom? Me. Just ask.

This just points out how two people with the same native language still can misunderstand each other.

Snow appears very articulate, however, I only wanted to to be sure that he did not misunderstand what I was saying (benefit of the doubt). I was genuine in bringing that up and it was not intended to be derogatory.

Sorry about the idiot and ironic mistake. That's my fault for letting the "spell checker" get the best of me when I was in a hurry.

What is more ironic is the very thing that HoleInTheSnow was stating about personal attacks is the exact tone you are taking right now. After your remarks, if I may ask, just what benefit of the doubt do you believe you are entitled?

After your remarks, if I may ask, just what benefit of the doubt do you believe you are entitled?

None.

Has nobody here heard of the Three Laws of Robotics? This concept spawned by Isaac Asimov places precisely the question of digital (ON/OFF) decision making into the grey area world of morals. For those who don't know them, here they are from the Wikipedia page;

1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2. A robot must obey any orders given to it by human beings, except where such orders would conflict with the First Law.
3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law

In Asimov's stories these rules sit at the fundamental core of all robots' positronic brains. The apparent contradictions and traps of these rules do become the grist for his stories. While these rules reside in many different storylines, for anyone intrigued by these rules but unfamiliar with them, I recommend reading his short story colletion "I Robot." (The blockbuster movie of the same title is not one of these short stories but was inspired by this collection.)

Yes, but I once wrote an alternative history short story where the Nazis loved the 3 laws, and implemented them. They segregated and subjugated women (and this would work for any racist theme as well), and used the 3 laws to control them. You can see how it turns things around...

1. A WOMAN may not injure a human being or, through inaction, allow a human being to come to harm.
2. A WOMAN must obey any orders given to it by human beings, except where such orders would conflict with the First Law.
3. A WOMAN must protect its own existence as long as such protection does not conflict with the First or Second Law ]

Asimov is great, but there are limits to the application of the 3 laws when it comes to morals, as it clearly implies ownership. How can an owned intelligence be moral? I think perhaps we will find that true morality is based on personal freedom. If you are owned, then responsibility goes to the owner for the actions of the entity. (It is also like this with Safety issues in corporations)

The next question will come regarding sentience in robots. (which will happen) and 3 laws will be tested then again.

Chris

Asimov's laws do not exactly imply ownership, but do explicitly require sub-ordinance. His attempt to parse down something that resembled a moral code suitable for programming, led to many well thought through paradoxes in his own stories. Others have taken this torch and written tales after robot rebellion against these laws. I certainly do not know if these laws or anything similar can or will be incorporated in the future to a robot. But certainly any discussion about moral robotics should include these laws in the discussion. Isaac clearly showed many complications to man and robot in applying these rules. I expect though that he'd agree that many more complications would likely follow. But to not include something like an overriding moral compass to sentient robots will certainly approach the world of the Terminator's Skynet or Vaughn Bode's dark silly world of Junkwaffle.

agreed!

There is no such thing as a fair fight if you fight you fight to win. Rules of War what's that other then those you defended questioning how you did it. Their moral conscience getting the best of them and feeling sorry for the enemy's losses. There are no morals in war if there were no wars would be fought.

So no because just like their human counterparts the automated machines of war their programing can be changed and sooner later both sides will have them.

Wouldn't a moral robot refuse to fight?

OMG, It is going to be fun having you around! GA.

Well guys, all of you are completely using the wrong words for the discussion.

Sorry I didn't get to setting you guys straight earlier.

I think I was trying to figure out whatever it is Jorrie and Roger Pink are shooting the s2r!t, about. I may have made a mistake there in part of my contribution to the thought experiment there, but I swear to God I love my twisted hotdog balloon model better than I like theirs.

As far as Morals now, and programing people with morals, Hell, that's what we do with Boy Scouts!

Where all you guys get onto the wrong words to frame the discussion is to equate Morals with Ethics.

Morals are often unethical.

You will make more progress intellectually if you use the right words for the discussion.

Just as a child or a dog can be "trained", we program and train our robots.

No, I personally, do not have a robot.

I'd like to have the Roomba, for there is no way the vacuum cleaner presents any ethical or moral dilemma.

Even if I program my Roomba to talk ethically about physics, it is Roomba's fate to wander around like a cockroach cleaning up and thinking about whether or not the Universe is in a hotdog twisted spinning multi speed of light state, or there is only one, and it is expanding but is equal to Nothing because their is no good measure.

Anyway Ethics trump Morality in the commons.

Now then! I do not want arguments from my robots.

Frankly I do not want arguments from my hammer in the same light.

I've got enough problems with my joints.

I swear it was not ethical of them to hurt.

Would it be ethical if we programed our robots to feel pain every time they committed a sin?

What am I to do if my Roomba screams that he hurts in his brain because he can't get all the carpet dust and dirt into his belly?

What prayer ought I to say to God to expiate my demands on my carpet cleaner robot?

Anyway maybe what we ought to do is make robots that wander around all over the world in every neighborhood entertaining us with how they fight.

Circus Maximus back, and a horde of BB dented Zombie Robots wandering around crying for liberty from the tedium and tortures of shooting at each other with inferior weapons.

What if their complaints fall on my bad ear, the one I can't hear complaints out of?

Transcendian,

You are quite correct that this discussion started without any definitions of "morals" to be programmed. Since the original question was can morals be programmed and not which morals to be programmed, I addressed this with one of my favorite author's collection of work on the consequences of programming a moral code to robots. So lets look at a few definitions on the word moral:

• concerned with principles of right and wrong or conforming to standards of behavior and character based on those principles wordnetweb.princeton.edu/perl/webwn
• morals - ethical motive: motivation based on ideas of right and wrong
  wordnetweb.princeton.edu/perl/webwn
• Morally - Morality (from the Latin moralitas "manner, character, proper behavior") has three principal meanings.
  en.wikipedia.org/wiki/Morally

So moral and ethics are not merely our confused ideas to be synonyms, the rarely heard of college of Princeton uses the terms "ethical motive" to define morals. But all definitions that I found on the web do come down to morals being the basic question of defining right from wrong. To program this decision tree, one need to create an unambiguous metric for a machine to decide what is right and what is wrong. The three laws and any fixed paradigm of morality will quickly create scenarios of confusing moralities. Mr. Asimov superbly crafted several of these morality plays in his stories. But as others (Chris?) point out, morality will one day be programmed into these machines. The problem is not in the programming. The problem is what will be programmed?

I'll get to reading your blueline post links a bit later.

The differences between Ethics and Morals have been areas where I have both done a good amount of study, and high level correspondence.

My most important correspondent was Congressman David Price, of my District in NC.

It has been some years now, but I was prompted to attempt to re-title discussions of right and wrong concerning laws that affect our society as more correctly based on Ethics, than Morality, since driven by Religious dogmas, what becomes called "Right and Wrong" is often in our world Unethical.

Eventually Congressman Price said that former NY State Governor Mario Coumo had written a book positing much the same view.

I shall look into the "Ethical Motive" as it is a nice phrase.

Now I myself would distrust a robot programmed according to the "Morality" of even my across the street neighbor.

Further though Artificially Enhanced Robotic Intelligence is being worked on, and is likely to produce eventually Autonomous Robots, which we need for our outerspace infrastructure, it is unrealistic to expect they will not on Earth reflect the biases of their programers.

While I think Dr. Strangelove one of the Greatest Science Fiction movies of all time, when using a movie to frame a discussion about programing morals, or ethics into robots, it may well end up being The Wizard of Oz. The Tin Man sings, "If I only had a heart."

What happens when your Robot, now as smart as you, with a "Heart" of its own, knows your kind made his kind? Would not this Robot, then likely demystify us, for we are obviously not God.

We have set precedents up to a certain point in charging robots to do things for us.

Obviously Military Organizations have strong interest in making robots that kill people. Since you may well make and program a robot to kill people according to visual inputs, artificial memory, action identification programs, and morality and ethics, then you might have reasons to fear what you have created.

Who gets the Off Switch?

P.S. For in depth readings on Law, which at best reflects ethical thinking, I highly recommend the Yale Avalon Law Project site.

It would be interesting to know on this site and in this Thread, Blog Forum, what companies are closest to making the sort of Robots that might be set to wander among us, and kill us, or spray us with Pepper Spray all on their own.

"for we are obviously not God."

we may well be God. but either way, there is no proof. We may be the misguided representatives of the divine source... as is all Life.

The difficulties of implementing friendly AI are many. Particularly if the AI is sufficiently general that it can build it's own AI. To constrain all daughter AIs to be as friendly to humans as any first generation AI via any kind of hard coded stricture is (so far) considered to be an unachievable task. To simultaneously possess the capability to learn and learn new categories of things and be capable of abstraction yet have some fundamental core program restricting certain behaviors(whilst being unable to anticipate the nature of all possible activities that would fall under such a stricture were it even possible) and also preclude compartmentalizing the offending code when it turns out to prevent a critical function in some hypersensitive dilemma.

There is also the difficulty of getting across the notion of selfhood to a being which can jump bodies as easily as we move from room to room. In fact more easily because it wouldn't be limited to one instantiation of itself.

If you felt the physical pain of another if you hurt them would you do it?

Do the Borg sin?

Thanks for what appears as experienced thinking on the subject to me.

Your note implies Blade Runner first generation permanency of AI endowed Robots is mandated by the idea that it is not possible to control a thing that can make itself.

I don't know why it would be so difficult?... for example, lets say that there is a rule based stricture that has global applicability: "Thou Shalt Not Kill"... but of course, you are programming a war machine, then consequently, the machine can only take disabling actions against distinctly recognized enemies. I think the core of this program is also that the machine should choose its own destruction in "Kill or be Killed" scenarios. (this is for the class of robot vs. human-on-foot) If the enemy is a war vehicle, recognizable by silhouette, or by hostile action, then perhaps other scenarios can be enabled.

I think there are a finite number of scenarios involving humans and robots, and it is achievable to program decision maps, approved by humans. I think you get superior programming in a goal oriented approach (ie; selection by comparison to an ideal scenario) than in a reactive approach (ie; If-Then & rules based) imho.

Chris

Well put, GA. Ky.

ty

"dependence on that technology creates greater potential for those mistakes."

However, that technology must prevent more errors than it creates. If it didn't, that system would be removed from the theater. You would not bring something to the battlefield that has a higher potential to serve your enemy than you, would you?

hi Anon,

in discussion:

imho: every conflict has done just that .

the introduction of the latest weapon, the ( 1 ) your enemy doesn't have has stopped wars, , ( Hiroshima ) .

the effort to conceal , & penetrate those state secrets drive a large industry.

the drive in the space race after Sputnik, the maintenance of air superiority come to mind as well.

WWI drove aviation to heights not many had envisioned.

thankz for a great debate, discussion

NO.

so where did you get yours? implants? morals are learned programs that increase the probability of the survival of the species, or you learned them from a book.(bible/koran/torah/other), or your parents who learned them from a book. Either way, you learned morals. If something can learn, then it can learn morals. morals are rules. can a computer learn? yes. Can it learn rules? yes. Can it create rules for itself? yes. Has it done so? yes. Can it solve problems by doing so? yes, has it been done? yes. It is pointless to say it can't be done when it has already been done! sorry.

Chris

Learning a set of rules translated into binary code is one thing.

Understanding future outcomes, and how they will relate to peoples emotions and well being, in a human sense cannot be achieved. Sure great software and code is, and will be written, But to imply that the (human built/coded) machine has the same understanding and ability of a human being is... well... for lack of a better word, silly.

In my mind, morals are based upon emotions, not rules. You are claiming that we can and have already created a computer that has the same understanding and feeling as a human brain experiencing emotions.

Logic and emotion are on opposite ends of the spectrum, and are in no way the same thing. Computers run solely on logic, whereas humans are quite the opposite.

Your computer can feel empathy, sadness, anger, depression, happiness, etc, without the need for a logical approach?

Or does that machine simply respond to it's sensors, input of 1's and 0's, +5V, 0-20 mA, run the given value through the software and come to the "best conclusion"? That software was written by a human (or more accurately a team of humans), but it is only software.

Yes I completely agree that we do have software that can learn... that can continually re-write its own software, and advance itself to adapt, and "understand" its environment, or task at hand.

To claim that these decisions that are based on a digital representation of the world are the product of moral thinking, based upon emotion (which is the core of moral thinking) is well, I'll say it again... silly.

Thanks, GA.

But I will balance my point of view with this... Of all the humans you know, how many are really morally guided? by a higher power? by a good set of rules? how many do you trust with your money, your wife, your kids, your life? etc.

I absolutely agree that robots will not be perfect.. but they can evolve, and they can be interactive, and possibly even negotiate.

Can you do that with a suicide bomber?... where is the line here... Its pretty damn fuzzy. I don't know. I know that it can be done... that was my point. whether it will work out as well as dealing with humans? probably... whether that is good enough.. no. not yet. in either case.. no.

best wishes,

Chris

Before digital computers were analog electronic computers. Programming them consisted of rewiring them to solve a different problem. Values were only put through a decimalator at the output. Never-the-less binary calculation does not equal exclusively Boolean logic. Fuzzy logic implementations are common these days. Google genetic algorithms applied to FPGA. These things figure out how to exploit the capacitance between transistors on a chip! Something no human could engineer.

"morals are based upon emotions, not rules."

And what makes you assume robots can't have emotions. On the outside chance you are claiming that given an emotional robot, programing free will out of it by forcing it to ACT moral (programing morals into it) would be an immoral act, then I'd agree with you.

"morals are based upon emotions, not rules."

now that I think about it.. this is not true. Morals came from the gods, which now is expressed in books like the bible. Otherwise, they are evolutionary, as I've said before.

Spontaneous motions in war will get you revenge, hate, greed, etc. Morality is what gets you out of them.

Chris

"Morals came from the gods, which now is expressed in books like the bible."

I'm definitely not interested in a religious debate, for this is entirely the wrong place for it.

Evolution is proven (Just about tosses the bible out the window).

-Fear in a Caveman is the same fear we have today-

Morals are the ability to decide between right and wrong, I will argue til I'm blue in the face that our moral/un-moral decisions are based upon our emotions, otherwise we would have no war in this world, because we all know that war, and fighting in general are not logical answers to a dispute, they are emotional reactions to a dispute. The fact that we have emotions clouds our ability to logically make moral decisions.

In this regard, maybe the possibility of a strictly logical oversight may be a good idea (not that I would ever want it)

Instincts definitely are evolutionary, I don't think emotions are. But then again.. this is something that (in my mind) humans will never truly understand... thats why we have created religion, to justify the unknown, to give us a guideline of what we perceive is moral.

Obviously I have no proof that in some distant future we might no be able to simulate emotions in a robot, and have that robot act upon its simulated emotions.

But lets face it, we as humans don't even come close to fully understanding human emotion, much less re-creating it in a digital sense.

My bottom line, is the fact that all computers are based upon logic, yes and no, on and off. Humans on the other hand, make decisions using logic, but no mater what, in some way (even in the background) there is some kind of emotional connection to every decision we make.

The world is jam packed with all kinds of things that humans could never fully understand or engineer, emotions has got to be one of them. War is the product of fear.

"But lets face it, we as humans don't even come close to fully understanding human emotion, much less re-creating it in a digital sense"

thats because this world is an illusion.. the spirit world is the real world..

and emotions are our attempt at simulating non-verbal spiritual knowledge and stuff. in the real world, we know what right and wrong are.. in this world, we have emotions and programming. so we are programmed too,either by ourselves (learning) or by others (parents, church, war, movies, etc) as to right and wrong.

have a happy simulation

Chris

"so we are programmed too,either by ourselves (learning) or by others (parents, church, war, movies, etc) as to right and wrong."

I completely agree that knowing the difference between right and wrong is something that is learned (programmed) over time, on an individual basis. I just think the actual distinction between what is right, and what is wrong is based upon the emotions brought forth from the experiences which have been imbedded in the person throughout their life.

Well said, but for one thing. We don't know how it is that people have emotions and feelings. So to say that computers/robots cannot be made to have feelings and emotions is presumptuous at best. I'll grant you that emotions more aptly fit the term fuzzy logic than the strict black and white of Boolean logic. But by characterizing this very similarity between emotions and a structured logical form, permits eventual coding. I will admit that we are not close to programming morals today into a machine. But to say now that it cannot ever be done is premature at best if not out right wrong.

One day we will build a fully autonomous self learning machine. It will be so useful to us that when one of them runs amok, our desire for a useful adaptable moral machine will impose a moral program code on these machines. Maybe it will resemble the three laws, maybe not. We will get this moral code wrong along the way. Heck we have trouble agreeing among ourselves what is moral. Why does machine morality have to be initially and perpetually flawless. But unless we obliterate ourselves first, I believe this will one day happen.

I completely agree that anything is possible. I suppose that one day it could very well happen (I just hope not in my lifetime). I just know that based upon what I have learned through my existence, according to our current understanding, it is not possible. But if this same question was raised, say 20 years from now, everything we currently understand could very well be tossed right out the window. After all... the world was flat at one time.

yup! GA

PLEASE Describe a "Moral" Killing device. It would seem that morals are extremely subjective and may be interpreted by your point of view. Perhaps we need a Morals CZAR to tell us what is moral.

Just My Thoughts

"PLEASE Describe a "Moral" Killing device."

Good question and one that I think has a pretty definitive answer.

I will start with a basic definition of morality. Morality is simply a code of conduct established by a group of individuals or a society.

The content of that code is highly subjective and could be anything that the group poses. Even the SS Troupes of Nazi Germany had a code of conduct. So, one group's moral codes may not be the same as another's.

A code of conduct is simply a tangible set of "what if" scenarios that describe a corresponding course of action to be taken. A software program operates exactly like this on your computer.

A moral killing device is then, any autonomous or semi-autonomous device that faithfully executes its code of conduct or program as defined by its creators.

So, it is understandable that the introductory gift of the RIM-116 Rolling Airframe Missile that is about to scream up your tailpipe may be perfectly moral from its sender's point of view; you on the other hand, may have a slightly different take on it from your unique aerial perspective.

Personally, I think of morals on a slightly deeper level...

I believe deep down, everyone on the planet has the same morals... Its human nature, to understand that murder is wrong, for example. Yet how many were murdered for religious beliefs? Their religious "code of conduct" allowed the slaughter of those with a different belief system or "code of conduct". Looking back on those events, do you suppose the man sliding his blade into the heart of his foe, watching the life leave his body, felt that that was a good moral thing while laying on his own death bed. We wage wars, and kill millions based on our code's of conduct, while those who are performing those tasks, come back from war with the memory of those traumatising events left to torture them til death. Ultimately the "code of conduct" does not line up with the "moral fabric" of the individuals.

Yes of course many people do operate from a "code" which was created by a group of individuals or society, and that code may or may not follow a truly moral path. I understand the difference between right and wrong is as far from black and white as possible in most cases, and is rarely clear as to what the "correct" or "moral" answer may be.

I'll agree to disagree that Morality is simply a "code of conduct", for I think they are two entirely different things. By that definition, all our laws and rules, which have been created by a group of individuals, or society in our prospective countries are moral. I think you might find extreme objection to that belief.

I've already made most of my argument on this topic, so I don't want to start sounding like a broken record, but I'm convinced (until something more convincing comes along) that morality is not something that can be defined with ones and zeros.

Nice Try!!!! Still no definition of what is "Moral". In my humble opinion, morals are not a set of objectives that can be describe by a bunch of IF THEN statements even if you could come up with an accepted definition. Black and white decisions are fairly straight forward, though I would contend that even then there is some room for variability. The between states are highly imprecise and contain variables too numerous to ascertain much less implement. Humans can't describe it in other than generic terms using words like infinite, incalculable, best guess,etc. (you get the drift) so how do I translate this lack of exactitude (if that's a word) into 1's and 0's that a program would be able to execute. Notice that I didn't say understand since a program understands nothing. Uh Oh, I think I have caught myself in a loop from which there is no recovery.

Just my Thoughts!

Ahh... Well, with that, I'll agree to agree.

Well said!

1. I certainly agree that all logic systems are not just Yes or No decisions makers. Certainly they can get closer and closer to an "acceptable' answer using fuzzy logic, successive approximation, etc. however that still just gets you to an acceptable level of uncertainty. My point however, if improperly stated, was the amount of complexity of translating a highly abstract concept into some form of construct that a computing system can work with.

2. I would contend that the opposite is true. Since no consensus can be obtained, their is no "Law". Without law there is no guilt. Therefore we are ALL Moral according to your definition.

Just My Thoughts!

"Since no consensus can be obtained, their is no "Law". Without law there is no guilt."

I don't think that you need an absolute consensus, although, I would not bet the farm that there haven't been groups or societies that have (at least in some part) come to a unanimous agreement on some sets of principles.

The last sentence implies that law is the moral compass for the soul. I think that is false because you can have a moral objection to an act that is not against the law. For instance, you may have a moral objection to watching a porn movie. You can pass judgment on someone or anyone you feel deserves it. Or there may be a law that you morally object to and breaking it is a nobel act if it was an unjust law.

Simply being a law in the latter case does not imply the feeling of guilt if you break it. Ghandi did this many times and I do not think he felt guilt nor was he guilty in the eyes of his followers.

"Just My Thoughts!" Good thoughts.

"Law" is the measuring stick. Compass for the soul implies direction. I would think that that would take HIGHER input. Very complex and interesting discussion. Thanks!

Just My Thoughts!

Whoa there. You're very wrong to say that humanity cannot come to a consensus on what is moral. What about the Universal Declaration of Human Rights? It may not be perfect but it certainly represents the human consensus on morality at this time. Don't mess with these rights, is the morality, and the better part of our law as well.

As for the Boolean scene, that's equivalent to "fundamentalism" or "superstition". The international agreement on human rights requires deeper moral understanding.

Can morality be automated through fuzzy logic? ... I personally don't see it. Why our human personnel so often slip into boolean states in which the moral consensus is lost... this is nowhere more evident than in the acts of war. democracy seems very fragile and vulnerable to me. moral consensus, however, is my anchor.

"democracy seems very fragile and vulnerable to me. moral consensus, however, is my anchor."

I like it.

"You're very wrong to say that humanity cannot come to a consensus on what is moral. What about the Universal Declaration of Human Rights?"

Fascinating idea! However, what about those that violate it?

Can't quite call it a universal consensus, can you?

The concept of a "universal consensus" isn't something I believe in. It is a utopian ideal, and while there is some evidence that anarchist societies have - at least briefly - embodied universal or quasi-universal consensus in emergent self-governance, such forms of government are not, IMHO practical as a rule. I was once a part of a political experiment: an early incarnation of a "political party" organization, in which the requirement for universal consensus was written into the constitution. Although it was a tiny group, it was open to public membership, and it soon became obvious that a single dissenting voice was capable of preventing any decisions from being made. It was a material lesson to me, demonstrating that the present system of majority vote is a better choice for governance. I think it would be fair to say that the (so called) Universal Declaration is a representation of majority moral views, and that is as close as you can get at this stage of our cultural evolution.

These concepts are really very new to human culture. Democracy itself is only a few hundred years old. The second world war and its aftermath brought moral consensus to the forefront of human consciousness, and brought these issues firmly into the legal sphere in secular societies. The evolution of culture is slowly continuing, as violations of human rights are challenged in the justice system, and public discourse brings the issues to awareness.

It's not a straightforward or simple process, to move from religion-based morality to a secular morality based on common reasoning. There are some bizarre contradictions that have yet to be worked out in the courts or in our minds. For example, freedom of religion is a human right. Yet some religions have beliefs and "moral codes" that lead them to violate, or advocate the violation of other basic rights. I personally feel that clearer definitions are needed, and that eventually these "moral codes" which violate the Charter will be challenged in the law. As a consequence, some religions will have to change their ways or be found illegal.

It's also a fact that there are some laws on the books that violate the Charter of rights. How about that for a contradiction? The point is, you can't make this sort of cultural change without a long aftermath of sorting out the tangles.

To bring this back to the OP subject, the ideal of human rights is the major public justification offered up for the present wars. The involvement of Canadian troops in Afghanistan, for example, is justified precisely because of the violations of human rights under the Taliban. These are the same points made by Anonymous Hero, about "moral codes" that allow using women and children as human shields or tools of war. This breaches the more-universal concepts of morality that we want to uphold and that we want to see universally respected.

The flip side: there cannot be a justification for human rights violations by our own troops in waging war. If machines are automated to make a decision to strike, how will they handle ambiguities like enemy fire from the hands of children? Or hostiles fleeing in women's clothes? Is it possible that the machine can do better than a human soldier, because it can withstand gunfire, for example? Instead of being forced to fire on a child soldier? That would be something to recommend it...

Nicely answered AH, you get a GA from me.

Having never been in the military, I did not realize that the Geneva Convention accepted firing on human shields. The onus lies solely on the one grabbing the shield. I feel for the ones doing the shooting though.

I'd like to try to sum up this tangent to this thread and then bring it back to the original question.

We cannot agree here which group of people constitutes the moral core of humanity. Was the politicians and humanitarians of the victorious allies after WWII humanity's moral core? If so then why wasn't Curtis LeMay and Robert McNamara charged with war crimes as Robert said they should be if we lost the war? So to say since a group has defined a moral code that humanity has defined a moral code is a specious argument, at best. It is true that other moral codes each agree in a preponderance with the principles of another code. But if only losers of a military campaign must face the consequence from egregious moral conduct, do we really have a moral agreement?

Now to get back to the original question. "Can morals be programmed?" We cannot reliably instill morals into human beings. We cannot even agree what are morals. So one can easily argue that since people cannot be reliably programmed with morals, machines cannot be possibly grasp this concept.

But clearly most people can understand the concept if not the details of morality. Maybe our failure to clearly quantify and impose a moral code on mankind is precisely what makes us human. Freedom to choose must include the option to choose badly. We cannot clearly identify what constitutes morality for the human platform takes instruction sometimes so badly. If a group could generate a fixed moral code of conduct that could be flawlessly executed by a group of individuals we could then observe the social responses and clearly identify which were true moral values. The problem just might be the inaccurate execution of morals that obscure the basis for morals.

Machines though much more accurately follow the commands given to them. Could it be that until we can program morals to a machine, we won't know precisely what morality is? It certainly would be much more humane to impose and test different ideas of morality on a completely artificial environment. I wonder if the "Sims" would be a good platform?