1
$\begingroup$

If the government were run by robots instead of politicians, and these robots had to reach out to the masses for every single decision they made, wouldn't the world be a better place because robots would be resistant to life threats by bullying parties to pass bad laws and engage in unwanted wars etc.?

The only downside is that keeping up with everything by citizens could take up a lot of their time, and they could engage in forums related to political issues and then vote a lot... But then what would be more important than politics and law in a "correct world"?

$\endgroup$
  • 1
    $\begingroup$ If the robots have to reach out to the masses for every decision, how is this different from direct democracy? I.e. why do you need robots? $\endgroup$ – Brythan Jul 9 '16 at 4:45
  • 1
    $\begingroup$ But why do you need politicians? It's a direct democracy. Politicians can't pass laws in direct democracies. What do the robots do? It seems that you are implying that they replace the politicians, but politicians are unnecessary for the system that you describe. So why replace them? $\endgroup$ – Brythan Jul 9 '16 at 5:05
  • 2
    $\begingroup$ Politicians are elected with campaign promises in mind. Then these promises run into a real world that is constantly changing. Somebody runs on the basis of lowering taxes and raising welfare, then there is a recession. Somebody else runs on the basis of changing abortion and gun laws. He doesn't get an outright majority, but there is someone else who would offer to help with one of the planks of the platform in exchange for the other. Politics is the art of forging compromise solutions. $\endgroup$ – o.m. Jul 9 '16 at 5:16
  • 2
    $\begingroup$ You should read the Revalation Space stories. I don't recall how much detail it goes into (I've not read the Prefect yet but it souds like he explores it in more detail here.) $\endgroup$ – JDługosz Jul 10 '16 at 21:31
  • 1
    $\begingroup$ You might also check the story mentioned at the end by Joan Vinge. $\endgroup$ – JDługosz Jul 10 '16 at 21:37
3
$\begingroup$

A perfect political AI would run a government perfectly. The problem is that perfect is really hard to define, and the stakes are quite high.

One approach might be a robot that follows rules. For example, it may be told to enforce the opinion of the majority. However, this is not as ideal as it may seem. The knowledge with which to make good decisions is not universal. The public is NOT educated enough to decide a location for a nuclear powerplant. The public is not educated enough to decide whether we should go to war. There isn't enough time in the day to be educated enough on all issues, this is why there is division of labor. Plus, in the case of war, some of the essential information may be classified, and not safe to dissiminate.

In general, rule based systems are hard to put in place well because you must ceede power to the rules. You lose the ultimate power the instant you engage the rules. If you made a mistake, well, rules are rules. You can add rules to change the rules but that only defers the problem. There is an infinite regression to deal with there.

On the other extreme, we can give the power to the AI to make decisions. However, how do we know those decisions will be in our interest before we enact the AI? There is more than one movie about a AI bound by the wrong rules. Some of them have even refused to open the pod bay doors. This ability to decide also opens the door for corruption. It will be in a slightly different form, rargetting an AId decisions rather than a human 's, but the effect would be the same

So what you want is a balance. The AI needs to have commitments to some rules, but the freedom to avoid catestrophic misinterpretations. You have to decide which things you want you AI to hold to while it is in office. You also have to get some agrements between all of the citizens. Everyone will have to have some say in what the AI will hold to.

Which is starting to sound a lot like campaign promises , don't you think?


Edit: from the comments, I mentioned Isaac Asimov's Three Laws of Robotics. 70 years after he wrote them, they're still considered some of the best rules ever written for a robot to follow when working with humans. For those who are not familiar with them:

  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 the orders given 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 Laws

Asimov then spent an entire career writing books about the loopholes in these laws, and how each one could play out. This just goes to show how nuanced such a program would have to be.

$\endgroup$
  • $\begingroup$ Complicated picture, but real enough. However, I think campaign promises should be coupled with education. I think that would make things better. Can you please comment on this observation? $\endgroup$ – Jack Maddington Jul 10 '16 at 22:43
  • $\begingroup$ Anyways, your explanation is outstanding. But does that mean there is no better system for us than the one we have today, because knowledge acquisition is time consuming, and once all people have the new knowledge, we're on done way back to where we started with? $\endgroup$ – Jack Maddington Jul 10 '16 at 22:46
  • 1
    $\begingroup$ @JackMaddington I don't think it means there is no better system, just that it's remarkably hard to find a better system with large sweeping changes. The system we have is as it is because it is balanced between a lot of conflicting goals. If you try to rewrite the system, it still has to strike a balance between those same goals. Now what would be interesting is trying to make a process to replace politicians with AIs slowly. The issue with the original plan is you need to get the answer right on the first try, when making the AI. If the process can be slow and subtle enough to... $\endgroup$ – Cort Ammon Jul 11 '16 at 2:04
  • 1
    $\begingroup$ ... give the voting citizens time to feel out how this change affects them, it would give them time to correct any initial mistakes. Start small, and let it grow =) $\endgroup$ – Cort Ammon Jul 11 '16 at 2:04
  • 1
    $\begingroup$ I'm certain each political party has several computer simulations to help them with such things, and I know I've seen the in games. Think tanks use them all the time. However, my favorite incubator for such ideas is Isaac Asimov's series of books about the Three Laws of Robotics. They were such good laws that they still stand tall 70 years later. I haven't heard of any rules that have been found to be "better." And yet, Asimov made a living off writing books about all the ways the Three Laws could fail. Each book explored a different loophole than the last. $\endgroup$ – Cort Ammon Jul 11 '16 at 14:42
4
$\begingroup$

Usually the term robot is used for machines with some sort of manipulators. What you are looking for would be an AI which may or may not be running on a robot; it could be running on a virtual machine in a cloud as well. But language use may be shifting in this regard.

Your plan that the machine would listen to popular opinion but not to "unethical" pressure is a contradiction. Most of the pressure groups also represent a slice of popular opinion, which may be very vocal even if they do not have a majority. If the machine can make ethical decisions about which pressure to resist and which pressure to accept as popular opinion, why wouldn't it be complex enough to worry about its own future?

Also, a society that is administrated without human compassion could be even worse than a society that is administrated with human passion.

$\endgroup$
  • $\begingroup$ I like your answer. Actually, I feel you have understood my question though, because many pressure groups which represent a large slice of society actually do lack in human compassion (think of assessing debilitating psychiatric drugs, unsuccessful cancer treatments, and large scale chemically toxic fertilizers and genetic modifications to foods, for instance). So, an AI agent would have to take the health and biological functioning of its people as Ann ethical principle to resist pressure groups. $\endgroup$ – Jack Maddington Jul 9 '16 at 5:17
  • $\begingroup$ But the pressure groups' propagandas and dissemination of false information prevention would have to be taken into account here. What is your take on this? What do you suggest must I do to improve my worldbuilding exercise while retaining the spirit of my original question? $\endgroup$ – Jack Maddington Jul 9 '16 at 5:18
  • 2
    $\begingroup$ You could have a fictional world where well-meaning people try to institute AI government, perhaps after a scandal. Make sure that some of the opponents of this scheme do so for good reasons. And have a few problems on the way. Perhaps the literal application of the law leads to a great injustice in a specific case. People call for the AI ruler to issue a pardon. The AI resists this pressure and the sentence stands. There is a disaster with GM food. People call for a ban on GM. The AI ruler resists this pressure and preserves the rights of GM companies. $\endgroup$ – o.m. Jul 9 '16 at 5:25
  • $\begingroup$ Then again, if the AI ruler were also a machine algorithm, so that it could resist life threats, how could it do these things, how could it being so "humanely smart" be plausible or even technically feasible from a computer science standpoint? $\endgroup$ – Jack Maddington Jul 9 '16 at 5:29
  • $\begingroup$ @JackMaddington, I don't think physical threats are the problem here. The most common assassins are lone madmen without a coherent agenda. Underhanded tactics come into play when corporations fund candidates for benefits, or fund both sides and then threaten to withdraw the money. An AI that is smart enough to rule a nation might be smart enough to want to stay in power, too, and make compromises. $\endgroup$ – o.m. Jul 9 '16 at 7:52
1
$\begingroup$

In Asimov’s story, the AI (named Multivac as with a number of stories that feature a computer) is so smart that it cannot be disobeyed, even thougn its global planning is only advisory and not law. If it recomended planting X acres of wheat, and you decided to disobey, well, it knew you would not all along, and secretly manipulated you to choosing what it really intended and/or included your predicted planting in its overall plans.

In another story, a Positronic robot passing as human ran for president, so it could do the most good for humanity.

In general, we have the problem that politicians are good at getting power, not necessarily good at anything else like rational problem solving. Having machines run things would be different in that they would be built for that purpose and very skilled in the necessary areas.

That would include compassion and selflessness, so it would not have the pressures you noted but neither would have the flaws noted by other answers. But really, it all depends on how the machine was made. In that sense the question has no definitive answer. I can answer as what could be optimal, but the story lies in how it really turned out.

$\endgroup$
  • $\begingroup$ Interesting answer. What I do not understand is how compassion can be built in. Often being compassionate towards one cause could mean being uncompassionate about something else. Also, it seems to me like language, which is what a robot would likely use as logic, would influence what was compassionate and what was not. $\endgroup$ – Jack Maddington Jul 10 '16 at 20:52
  • $\begingroup$ People like the Multivac in Asimov's story exist in real life in my opinion. But they are the smart people nobody knows who actually are able to influence or rule elected politicians, lawyers, and other people in positions of power. $\endgroup$ – Jack Maddington Jul 10 '16 at 20:55
  • $\begingroup$ How would a machine know how someone felt? If such person did not report back the machine might never know. But without doing something to that person they might not know either, and that something could be somewhat uncompassionate? How could the machine possibly know every action',s outcome if programmed by ignorant people? $\endgroup$ – Jack Maddington Jul 10 '16 at 20:58
  • $\begingroup$ Know how someone feels: how does anyone? Multivac knowing enough to predict behavior: not explained. today can suppose global detailed information gathering like in the TV show Person of Interest. $\endgroup$ – JDługosz Jul 10 '16 at 21:38
  • $\begingroup$ I've never seen the show "Person of Interest", must be interesting. Do you have a YouTube link? $\endgroup$ – Jack Maddington Jul 10 '16 at 22:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.