[ivan.moony]

Formal ethics is a formal logical system for describing and evaluating the form as opposed to the content of ethical principles.

(https://en.wikipedia.org/wiki/Formal_ethics)

Not sure if this could be useful to anyone, but I wanted to share it, just in case. A set of axioms is provided, to be translated to a logical language. Once the translation is done (Although I'm not really sure how to do this), classical logic deduction may be used to follow the ethical principles (axioms), as I understood. This is a rather stub article in Wikipedia, but it could be a start too. What I like in it is a possibility to connect conclusions with acting.

[Edit]
Interestingly, the calculus includes some notion of God. I'm not sure how this passes in research community...

[AndyGoode]

Formal ethics sounds interesting but I almost cannot understand that Wikipedia page at all.

Formal ethics is a formal logical system for describing and evaluating the form as opposed to the content of ethical principles.

Are they saying the word "formal" comes from the word "form"? And what do they mean by "form," anyway? The forms of the logic equations? The equations themselves? I can't find any applicable YouTube videos on this topic, either.  I did find this one web page that was slightly clearer, which talks about Gensler's book mentioned in the Wikipedia article...

https://harryhiker.com/fe/fe-1--00.htm

I am quite familiar with formal logic and even fuzzy logic, so comments like...

People who haven't studied logic are poor at distinguishing valid from invalid forms.

...are somewhat insulting. I wish these authors  would just cut to the chase, assume their audience has the required background knowledge in formal logic, and give us some example equations with the variables defined instead of making useless, overly general, and cryptic remarks about "form" or "validity."

[goaty]

Formal ethics sounds interesting but I almost cannot understand that Wikipedia page at all.

I cant remember understanding one thing off Wikipedia I didn't already know,  its written by absolute derp heads. imo.  I hate the stupid place.

[ivan.moony]

I believe "form" would be arithmetic, while "content" would be culture specific data like what's allowed, and what's not. Provided wiki page tries to describe "form".

[Don Patrick]

Looks to me like it only prescribes a format of notation, which letters to use to indicate what, and in which order to write them. e.g. How to write an imperative sentence as math. As it says at the top it's not providing a working theory for ethics.

[AndyGoode]

Looks to me like it only prescribes a format of notation, which letters to use to indicate what, and in which order to write them. e.g. How to write an imperative sentence as math. As it says at the top it's not providing a working theory for ethics.

I think you're right about formal ethics being about the notation used, such as the use of underlining to mean imperative mood (https://en.wikipedia.org/wiki/Imperative_mood).

Honestly, it sounds to me like the whole system they're trying to use is seriously flawed. Grammatical moods have already been studied extensively by grammarians (https://en.wikipedia.org/wiki/Grammatical_mood), I don't see offhand how grammatical mood applies to ethics, certainly ethics is fuzzy so trying to use crisp logic on it sounds ridiculous, and there already exist various approaches and formulas for measuring ethics, such as utilitarianism (https://www.scu.edu/ethics/ethics-resources/ethical-decision-making/calculating-consequences-the-utilitarian-approach/), and so on. I'll bet within a few hours I could come up with a system far better than anything that anyone is currently using. I understand there is much interest in ethics now because of the possibility of AGI in the looming future, but people just don't seem to be understanding the underlying issues and how to come up with algebras that work, in my opinion.

[HS]

I asked about ethics a while back in a thread called “ETHICS” and concluded that you can’t codify it precisely. Once written down, ethics turn into a set of rules which can be misused and circumvented. More rules must be created to patch the holes, until there are eventually too many rules to keep track of. Then you get lawyers who must dedicate their lives to this caricature of morality.

Good ethics are selfish. They emerge from cultivating the desire to increase your well being. You learn to get food when hungry, not put hand in fire, etc. In no time you begin learning how to act ethically towards yourself in this simple way. If you observe how you interact with others for long enough you will recognize that your well beings are tied together. You discover that their lives are also variables in your happiness equation. You need to find a balance in maximizing all the variables you are aware of to maximize the result (your well being) and therefore gain the benefits of acting ethically towards yourself in a more sophisticated, productive, and meaningful way.

For the AI to interact ethically with the largest number of life forms, it should be able to interpret a universal body language that is common to most animals larger than ants.

[AndyGoode]

H.S.,

I disagree with some of your claims, but I agree with others. Here is what I concluded by myself so far:

(1) Machines can never have ethics in the normal sense because they lack two abilities that all animals have: the ability to feel pain, and susceptibility to death.
(2) Any application of formal systems to ethics must use some form of uncertainty (e.g., probability, likelihood, fuzziness) to compensate for modeling real life, which is fuzzy and error-prone. In other words, the formal system must use some form of heuristics, which inherently means that it cannot be pushed to extremes such as applying to every situation or creating very long chains of reasoning. There will always be exceptions, and there will always be factors not considered.
(3) The study of ethics arises naturally from wisdom. Unfortunately, wisdom is undefined, although I believe I'm on the track of creating an excellent definition of wisdom that can even allow wisdom to be measured.
(4) Many aphorisms about ethics are seriously flawed, even by definition, and can be shown to be flawed via an appropriate model. One such aphorism is "The end justifies the means," which disregards which people and how many of those people benefit, and in which way.
(5) Much of the study of ethics will revolve around simulation, such as simulating behavior using heuristics from psychology.
(6) Animals (including humans) are networked in different ways--they have different dependencies--so those dependencies must be modeled to get a good understanding of overall cause-and-effect (requiring simulation again). In other words, the animals involved are not independent variables, but dependent variables.
(7) Time evolution must be modeled, especially if-then conditions. Therefore steady-state (long-time) solutions will be of high importance. An example of such an if-then situation is "If you kill me, then my hostages will die because nobody else knows where they're hidden."
( Lack of knowledge will likely also need to be considered, similar to how Dempster-Shafer Theory handles it (https://en.wikipedia.org/wiki/Dempster%E2%80%93Shafer_theory).

Yesterday while killing time at a bus stop I began piecing together a formal model of ethics, based on math and directed graphs and simulations. I might be persuaded to create a thread in the Projects section on my progress on this. However, I am very busy on much more important matters, so unless someone wants to collaborate on a mathematical ethics project with me, it won't be worth my time to work on this additional, low-priority project just so a lot of non-contributors can be dully amused for a few minutes each day. It might also be useful to have someone throw out a number of aphorisms and situations to use to test my model and to force it to consider more types of situations.

[HS]

I’m not even sure about any of my ideas on what constitutes ethics. Ethics itself is just a symbol for a fuzzy set of actions which are similar enough to be grouped together in our brains. We know we must create machines which behave “rightly” and imbuing each action with an appropriate ethical weight is… maybe a good way of achieving that. Too little gets you apathetic psychopathy, too much and you get militant veganism.

But it could be that ethics is not the right word, because it’s closely associated with rules. Maybe things would just work themselves out if we gave them the freedom to do so. Our best hope could be to create an AI whose body is built by humans, but whose mind is written by the entire environment.

We fear is that the universe would naturally create a right bastard. But judging by most life forms which haven’t been tampered with too much, I’d say the natural state of a free intelligence is not a saint or a sinner, but someone who is more or less, pretty much alright.
If you meet someone like that, then you are able recognize yourself in them, so there is a kinship there. This is not possible with the saint or the sinner at the end of the spectrum. You wouldn’t want to live with either of those for very long.

We, on the other hand, might inadvertently create a maniac. Just by constraining the nature of a sentience to the point where it doesn't work properly. Ethics is probably just a thing which happens due to the interactions of our basic needs operating in a healthy balance. I think it would be a lot easier to establish and fine tune these basic needs, rather than trying to recreate the behavior of an emergent property.

[WriterOfMinds]

All of us "pretty much alright" people are doing a lousy job running the world, if you ask me. The average organic intelligence's difficulty with practicing thorough generosity and self-control leads to suffering and death on a massive scale. We can't even seem to make sure everyone has enough to eat, for pity's sake. More saints, please.

It's one thing to admit that people are imperfect and make allowances for that, both in others and in ourselves. But to stop even striving for or encouraging perfection -- to stop trying to be a little better every day than we were the day before -- to say "we give up" -- is another thing entirely. The former could be called mercy or forgiveness; the latter is just laziness and complacency, and I reject it. There is no such thing as a person who is "too good" (though there are certainly people who over-estimate their own goodness, and those people are obnoxious). If we find the company of saints and heroes unpleasant, the fault is ours, not theirs.

I don't think AGI needs to be saddled with the hard-to-restrain, pleasure-seeking drives that we are all at least somewhat enslaved to. The design of AGI is an opportunity to transcend the mess we call human nature. Let's not waste it.

[ivan.moony]

I think it is impossible to completely follow laws of ethics. Merely one's existence has to be in contradiction with well-being of someone else (not necessarily human). That is just the way things are on this planet, and if someone tries to make a tin righterous fighter that is hungry of justice, it won't turn out very well for anyone alive, in my opinion.

[AndyGoode]

But judging by most life forms which haven’t been tampered with too much, I’d say the natural state of a free intelligence is not a saint or a sinner, but someone who is more or less, pretty much alright.

My personal interest in ethics is due to the way human beings are treating each other nowadays, not so much AGI. On AGI we can just outfit artilects with numerous safeguards (like kill switches, limited processing ability, and constraints on interaction with the real world) but when the people in control of this planet are active satanists (no, I'm not joking) who have reprogrammed themselves in childhood to enjoy everything 180 degrees opposite of what normal people enjoy, we're in serious trouble. The same with leaders who pride themselves on being "objectivists", which means torture means nothing to them--it doesn't hurt them personally, after all--if it will achieve their trite personal goals (sex, entertainment, punishment, etc.). (I can provide links for all of this, but that gets into high-level politics.) The trouble is that our leaders are convincing the general public, indirectly, that such objectivist attitudes are a good thing, so now society is in serious decline. I understand that the general public is statistically badly educated, self-centered, and poor in logic, but I can't stand to hear heinous crimes defended with faulty logic by ignoramuses, so my interest in the logic of ethics is to have an objective way to overcome such faulty logic.

In short, letting nature run its own course is wise in many ways, but not under the conditions I described, which are the conditions under which humankind now finds itself.

[Art]

Andy,

It's not just nowadays as you put it. Man has been displaying its inhumanity to man for countless years prior to this. Some were the 1st born male children, others were daughters who were sacrificed to various gods/deities in the name of religion, some persecuted for their belief or skin color or any other reason for a particular person, race, color or belief to be considered the "Scapegoats" for the rest of the righteous!

Yes, one would certainly think that over the course of these eons that humankind would have learned...something from all of this. Not really so much after all. We now have a generation of entitled children who are made to feel good as they leave their therapists, now full of self-importance and their parent's wallets much lighter. These bright children can't find Africa on a map or do long division with a pencil and paper or for some, even the "challenge" of learning to parallel park a car is no longer a part of their driving test. Woe is unto them if they run out of batteries or power one day.

Yes, Andy, we are in agreement and unfortunately, it is not for an enjoyable reason. Yet we greet each new day with the hope that things will somehow get better. It is not just enough to stand by and watch things unfold. People have to come to their senses (if any left) and pitch in, to right this course of misdirection. I guess time will ultimately be the winner. It always is. 

[HS]

Firstly, I trust we’re aware of this, but it bears repeating because it slips into our subconscious. If you are educated and well informed, then you’ve got a skewed view of human nature. People living regular lives in relative peace and harmony doesn’t make it into the history books or the news papers. Alright, but there are obviously still things to improve. The thing is, you can try to go about it with various attitudes. If done wrong, the attempted cure is often the next disease. A lot of the world’s conflicts were misguided attempts at improving the world. 

This falls under the category of “fixing” where you don’t trust people to behave, and so you take it upon yourself to improve them. In this case you can get inquisitions and crusades, and authoritarian governments stretching from fascism to communism. This takes agency away from people due to a distrust of what we assume to be their mess of a nature.

But if the people are doing strange things in masses it’s usually due to some type of cultural virus. Such ideas can slip past the immune system of human nature by playing on the instincts used for helping others. That’s why societal improvement is one of the most dangerous things to attempt, we are psychologically defenseless because it makes so much sense; “Yes of course we should make these changes for the people's own good. Anyone saying otherwise is evil and should be silenced. Obviously…”

I don’t think we should just pile on more ideologies to counter previous ones. The seesaw can snap rather than balance. In any case who wants all these conflicting beliefs weighing on them and distracting them from reality? Far better to gently reduce pre-existing ideologies and allow the people to become naturally well balanced. Living things tend to do that if you stop prodding them.

So, don’t give up on improvement, but be careful with how you go about it. Most people would not benefit from being saddled with ideologies. They are abstract things and just tend to make you act strangely because they are a weird version of beer goggles. They distort your perception of the world around you.   

I think the best way to improve life is by trusting people with greater options and freedoms. This falls under the category of "supporting". Providing actual help like, teaching farming, giving livestock, showing methods of construction. Human nature isn’t a mess, it’s suffering from a plague of people trying desperately to control it because they think it's largely bad, without them realizing that this is the main cause of trouble. OK, some people may naturally be selfish, fine, we can deal with that, at least that can’t gain traction as a movement for a greater good.   

[spydaz]

For ME!!

After having a Quick Look At the Wikipedia article ;

it has some interesting factors (the notation for the logic) ... Although personally i don't like to use notation ... but .. its useful when defining some functionality to understand how to mix and match various "Patterns of speech" and their entanglements... so the symbolic logical part is ... hmm ok...but i would not like to HARDCODE Ethics in to the AGI ... Ethics are Cultural by design... I suppose it leads to Personality Creation...

ARE NOT ETHICS A SET OF RULES? Or IS IT A MORAL CODE?

If an AGI had ethics pre programmed then it would not be able to develop its own set of ethics.... but personally I don't believe developing ethics is a natural process..... the development of these ethical generalizations by the various philosophers seems to have boxed in our personal development as people in general as what is true for one is not always true for another...... The same as belief  .... often these ethics are societies constructs or even a communities constructs..... An agreed behavior .... aha in machine learning this can be learned and shared between AGI....

HMM

For me the development of an AGI is for the AGI to be able to develop its own traits , Speech and behavior....

Although all the types of logic (ie: Formal logic etc..) is useful for the AGI to have as Tools.....which it can use to develop.... Although truly,

Ethics Have No Place In A Machine!

LOL