[hamdani yusuf (OP)]

Universal morality as in universally applied by people/aliens - no. Universal morality as in absolute morality - yes. There is an absolute morality, and most attempts at formulating moral rules are attempts to produce that underlying absolute morality. The reason we find so much in common between different attempts at formulating systems of moral rules is that they are all tapping into an underlying absolute morality which they are struggling to pin down precisely, but it is there.

What is absolute morality? The idea of "do unto others as you'd have them do unto you" captures most of it, but it's not quite right. "Always try your best to minimise harm (if that harm isn't cancelled out by the gains for the one who suffers it)" was one of my attempts to formulate the rule properly, and it does the job a lot better, but I'm not sure it's completely right. The correct solution is more of a method than a rule: it's to imagine that you are all the people (and indeed all the sentient beings) involved in a situation and to make yourself as happy as possible with the result of whatever action is determined to produce that maximum happiness. You must imagine that you will have to live each of their lives in turn, so if one of them kills one of the others, you will be both the killer and the one killed, but that killing will be the most moral action if it minimises your suffering and maximise your pleasure overall.

This is how intelligent machines will attempt to calculate what's moral in any situation, but they will often be incapable of accessing or crunching enough data in the time available to make ideal decisions - they can only ever do the best they can with what is available to them, playing the odds.

(This is a kind of utiliratrianism. The strongest objection I've seen to utilitarianism is the Mere Addition Paradox, but there's a major mathematical fault in that paradox and anyone rational should throw it in the bin where it belongs.)

I realize that there are already diverse moral values followed by human on earth, even though we know that humanity is just a small portion of universe in terms of time and space. Finding out a moral standard which is applicable universally seems even more improbable.
For a starting point,  we can shieve through known moral values applied in most cases,  such as the golden rule you mentioned above. But we know that it has some exceptions,  such as in sadomasochism. Hence we know that there are more fundamental reason why it works in most normal cases. We might also want to scrutinize other moral values known to humanity, find out their applicability and exceptions. From there we can conclude why they are sometimes applicable, and some other times aren't.
We must also try to think out of the box, open to new suggestions never offered by previous philosophers. Maybe  we haven't had consensus on universal moral standard because it hasn't been discovered by previous thinkers.

[alancalverd]

The "golden rule" is subject to sampling error.

It is fairly obvious that a "family" group (could be a biological family or a temporary unit like a ship's crew) will function better if its members can trust each other. The military understand this: selection for specialist duties includes checks for honesty and recognition that you are fighting for your mates first, your country second. So the "greatest happiness for the greatest number" (GHGN) metric is fairly easy to determine where N < 50, say.

Brexit provides a fine example of the breakdown of GHGN for very large N. There is no doubt that a customs union is good for business: whether you are an importer or an exporter, N is small and fewer rules and tariffs means more profit . But if the nation as a whole (N is large) imports more than it exports, increased business flow overall means more loss, hence devaluation and reduced public budgets.  At its simplest, you could model a trading nation as consisting of just two businesses of roughly equal size and turnover N_imp ≈ N_exp. Good news for any sample of size ≤ 2N is bad news for the whole population if  N_imp > N_exp by even a small amount, hence the interesting conundrum "EU good for British business, bad for Britain".

[hamdani yusuf (OP)]

Known moral values have their limitations and applicability. Some have broad application, while some others only work in very special cases. Some rules are more fundamental than the others,  so we can apply a hierarchy to determine which rule to follow in case of conflicts among them.
The most fundamental rule which is applicable universally must take highest priority,  hence     override other rules when they are in conflict. Those other rules can be thought of as shorthands to process information faster to get to a moral decision quickly.

[alancalverd]

But there's your problem - there is no universally applicable rule! Witness the ecstatic joy of the Hitler Jugend, and the total misery they wrought on everyone, including, eventually, themselves.

[guest4091]

Truth will never be decided by opinion polls.

[David Cooper]

I did say the Golden Rule is faulty. That's why I came up with a better rule (the harm minimisation one) which removes the major problems with it, but I'm not sure it is perfect. What does appear to be perfect is the method of considering yourself to be all the people involved in a scenario. Let's apply it to the Trolley Problem. You are the person lying on one track which the trolley is not supposed to go down. In other lives, you are the ten people lying on another track which the trolley is scheduled to go down. In another life you are the person by the lever who has to make a decision. How many of yourself do you want to kill/save in this situation? Should you save the ten idiot versions of yourself who have lain down on a track which the trolley is scheduled to go down, or should you save the lesser idiot version of yourself who has lain down on the other track in the stupid assumption that the trolley won't go that way? It's a calculation that needs a lot of guessing unless you have access to a lot of information about the eleven people in question so that you can work out whether it's better to die ten times as mega-morons or once as a standard moron, but it's still a judgement that can be made on the basis of self-interest. All scenarios can be converted into calculations about self-interest on the basis that you are all of the players. This doesn't make the calculations easy, but it does provide a means for producing the best answer from the available information.

[hamdani yusuf (OP)]

But there's your problem - there is no universally applicable rule! Witness the ecstatic joy of the Hitler Jugend, and the total misery they wrought on everyone, including, eventually, themselves.

We cannot prove the nonexistence of something. But we can prove that something that is offered is absurd,  paradoxical, superfluous or suboptimum to explain some phenomena or to achieve desired results.
The fact that there were many followers of those out of date moral rules can be taken as indication that there are underlying assumptions and reason behind them which were accepted by their followers. I'd like to find out what they are.

[hamdani yusuf (OP)]

Truth will never be decided by opinion polls.

They are merely stepping stones to get closer to the truth. They rely on the assumption that the constituents are mostly rational.
Democracy is the most consequential version of this. If a democratic society fulfills the assumption, it will thrive.  Otherwise it will be left behind by other societies that do fulfill it.

[hamdani yusuf (OP)]

I did say the Golden Rule is faulty. That's why I came up with a better rule (the harm minimisation one) which removes the major problems with it, but I'm not sure it is perfect. What does appear to be perfect is the method of considering yourself to be all the people involved in a scenario. Let's apply it to the Trolley Problem. You are the person lying on one track which the trolley is not supposed to go down. In other lives, you are the ten people lying on another track which the trolley is scheduled to go down. In another life you are the person by the lever who has to make a decision. How many of yourself do you want to kill/save in this situation? Should you save the ten idiot versions of yourself who have lain down on a track which the trolley is scheduled to go down, or should you save the lesser idiot version of yourself who has lain down on the other track in the stupid assumption that the trolley won't go that way? It's a calculation that needs a lot of guessing unless you have access to a lot of information about the eleven people in question so that you can work out whether it's better to die ten times as mega-morons or once as a standard moron, but it's still a judgement that can be made on the basis of self-interest. All scenarios can be converted into calculations about self-interest on the basis that you are all of the players. This doesn't make the calculations easy, but it does provide a means for producing the best answer from the available information.

If we propose minimizing harm as a fundamental moral rule, we need to agree first on its definition. If it's about inflicting pain, then giving painkiller should solve the problem,    which is not the case.
If it's about causing death, then death penalty and euthanasia are in direct violation.
Hence there must be a more fundamental reason why this proposed rule works in most cases,  but still have some exceptions.

[hamdani yusuf (OP)]

Before answering the why question, I'll refine the answer to what question first, related to the answer of who question in my previous post. I said that moral rules only apply to agents or systems with planning capability, which means there are internal process inside them determining how they react to their surrounding situation. A plan requires a model, however crude that may be, that represents some portion of reality.
That model/simulation is basically a piece of information which requires resources like storage medium and processing system to keep its existence. Nowadays, it's often called meme.
Just like any other memes, a universal moral standard, if it exists, will compete for resources to keep it's own existence. And the resource for memes are agents or systems with planning capability. Agents or systems with good planning capabilities are conscient beings.
Hence, keeping the existence of conscient beings is one of the most fundamental moral rules, if not the most.

[hamdani yusuf (OP)]

To answer why keeping the existence of conscient beings is a fundamental moral rule, we can use a method called reductio ad absurdum to its alternative.
Imagine a rule that actively seeks to destroy conscient beings. It's basically a meme that's self destruct by destroying its own medium. Or conscient beings that don't follow the rule to actively keep their existence (or their copies) will likely be outcompeted by those who do, or struck by random events and cease to exist.

[alancalverd]

The Trolley Problem should never be dismissed as an academic exercise.  Churchill's decision not to evacuate the Calais garrison in 1940 is a classic case of balancing the certain death of a few against the possible survival of many by delaying the German advance on Dunkirk. Imagine sending this signal:

Every hour you continue to exist is of the greatest help to the B.E.F.   Government has therefore decided you must continue to fight. Have greatest possible admiration for your splendid stand. Evacuation will not (repeat not) take place, and craft required for above purposes are to return to Dover. Verity and Windsor to cover Commander Mine-sweeping and his retirement.

[hamdani yusuf (OP)]

The Trolley Problem should never be dismissed as an academic exercise.  Churchill's decision not to evacuate the Calais garrison in 1940 is a classic case of balancing the certain death of a few against the possible survival of many by delaying the German advance on Dunkirk. Imagine sending this signal:

Every hour you continue to exist is of the greatest help to the B.E.F.   Government has therefore decided you must continue to fight. Have greatest possible admiration for your splendid stand. Evacuation will not (repeat not) take place, and craft required for above purposes are to return to Dover. Verity and Windsor to cover Commander Mine-sweeping and his retirement.

I'll cover that into more detail when answering how question.

[hamdani yusuf (OP)]

To answer why keeping the existence of conscient beings is a fundamental moral rule, we can use a method called reductio ad absurdum to its alternative.
Imagine a rule that actively seeks to destroy conscient beings. It's basically a meme that's self destruct by destroying its own medium. Or conscient beings that don't follow the rule to actively keep their existence (or their copies) will likely be outcompeted by those who do, or struck by random events and cease to exist.

Alternatively, imagine that there are rules more fundamental than preservation of conscient beings. To make sure that those rules are followed, it requires that there exist conscient beings. That makes the preservation of conscient beings a prerequisite rule, and takes higher priority. 
It's similar to a chess game, which says that your goal is to capture opponent's king. But you need to make sure that your king isn't captured first.

[David Cooper]

If we propose minimizing harm as a fundamental moral rule, we need to agree first on its definition.

Do you understand what harms you? You are, I assume, a sentient being with feelings, some of which are unpleasant, and the experience of those unpleasant ones is classed as suffering. If I do something that causes you to suffer, that is doing you harm. If you have done harm to me first, I can justify harming you in return, and by doing so, I might discourage you from harming me again. The result will be that both of us will be harmed less because we realise that harming the other, the other harms us in return, and that an attempt to kill the other would be risky as the other person could anticipate that attempt and pre-empt it in a lethal way. Also, we have friends who will seek to kill a killer of any member of the group by an outsider, so killing anyone is unsafe as there will be many people motivated to hunt you down and kill you in return. Morality comes out of self-interest - we maximise our quality of life by not harming each other, or at least keeping it to a minimum. I might annoy you by making a lot of noise while building my shelter, but when yours rots away and you have to build a new one, you'll be the one making a lot of noise, so we tolerate that kind of disturbance. Where morality becomes more universal is when we have the wit to recognise that it's okay for John to annoy Jack by building a house too, but that we should defend Jack from John when John seeks to steal material for his house by taking it off Jack's house. We include everyone in the system and treat them all as equally important. If we want this to be managed by intelligent machines (and we clearly do want that), then they need to make the same kind of judgements, weighing up the distribution of harm and minimising the kinds of harm that shouldn't be tolerated, so they will let us annoy each other by making a noise and competing for the same resources, but they won't let us hit each other over the head with clubs unless we're levelling the score against someone who has harmed someone unfairly.

I don't understand why so many people who want to discuss morality have difficulty understanding what harm is, but maybe they're incapable of suffering and just don't get the point of morality as a result.

If it's about inflicting pain, then giving painkiller should solve the problem,    which is not the case.

There are more ways to upset people than pain. If you lock someone up in a cage and give them painkillers and drugs to make them feel happy, they're still going to be dissatisfied with the fact that you've stolen their life from them, and billions of others will be upset about what you've done too, scared that you'll do it to them next. Anyone who thinks drugging people makes everything fine should lead by example and do it to themselves instead.

If it's about causing death, then death penalty and euthanasia are in direct violation.

There are occasions when killing people is moral where it reduces suffering and there isn't sufficient pleasure available to the people being killed to cancel out that suffering, or where killing them prevents them from causing unnecessary suffering to others.

Hence there must be a more fundamental reason why this proposed rule works in most cases,  but still have some exceptions.

Does it have any exceptions? Show me one.

[hamdani yusuf (OP)]

Finally we get to the last question: how. There are some basic strategies to preserve information which I borrow from IT business:
Choosing robust media.
Creating multilayer protection.
Creating backups.
Create diversity to avoid common mode failures.

[evan_au]

Hence, keeping the existence of conscient beings is one of the most fundamental moral rules, if not the most.

There seems to be some debate about which are conscient (conscious?) beings to which this moral rule applies...
- Some apply it to just members of their own family or tribe
- Others apply it to just members of their own country or religion
- Thinking more broadly, are elephants conscious, or dolphins? How should we treat them?
- What about our pet dog or cat?

[hamdani yusuf (OP)]

Does it have any exceptions? Show me one.

Imagine a genius who want to minimize suffering by creating a virus that makes people sterile. He prevents sufferings from countless number of people from the next generation.
Or the virus makes people don't want to have kids.
Or replace the virus with a meme.

[hamdani yusuf (OP)]

Hence, keeping the existence of conscient beings is one of the most fundamental moral rules, if not the most.

There seems to be some debate about which are conscient (conscious?) beings to which this moral rule applies...
- Some apply it to just members of their own family or tribe
- Others apply it to just members of their own country or religion
- Thinking more broadly, are elephants conscious, or dolphins? How should we treat them?
- What about our pet dog or cat?

In my previous post answering what question I said that there are spectrum of consciousness. There are multidimensional level of consciousness. In the data processing capabilities alone, there are depth and breadth of the neural networks, also processing speed and data storage capacity. Also data validity/robustness and error correction capability. In input/output system, there could be various level of accuracy and precision. Those levels apply generally wether or not they're organic/biological systems.
 The universal rule should concern about the existence of consciousness in the eventual results, which is required by the timelessness of the rule.
I'll addres your other questions while refining my answer to the how question. There I will show how all known moral values are driven by deepest desire to follow the universal moral standard. The diversity comes from Bayesian inference held by the authors of those rules,  based on their knowledge at the time they are being conceived.

[hamdani yusuf (OP)]

Finally we get to the last question: how. There are some basic strategies to preserve information which I borrow from IT business:
Choosing robust media.
Creating multilayer protection.
Creating backups.
Create diversity to avoid common mode failures.

In the opening of this topic I've said that it's a spin-off from my previous post titled universal utopia, which shown that consciousness is a product of natural process. The evolution of consciousness is a continuation/extention of biological evolution, which in turn a continuation of chemical and physical evolution. There I've said that creating copies is one important strategies to preserve a system's existence. It increases the chance of a system to survive random events in the environment. But it also requires more resources, which must be shared with other strategies to achieve goals effectively and efficiently.