This topic explores the implications of AI implementation of evolutionary law
Dana Edwards has been researching approaches to the use of AI in future law & it is an ongoing discussion in a different forum. One idea that interests me is an evolutionary law methodology, where AI constantly captures data on laws existing, laws passed, legal actions & court cases, & impact on individuals & contractual interactions.
The legal AI would evaluate the data, determine inconsistent results, & propose or implement changes.
Mark Larkento To clarify the polycentric law I was/am researching applies only to virtual space(such as running on Ethereum). I’m not so sure that AI could be of benefit outside of virtual space to the current legal processes unless voters say they want algorithmic polycentric law and most don’t even know what it is.
My opinion is that polycentric law belongs in virtual space. It can be tested in virtual space without the risks you would acquire by trying to take the option you mentioned above.
Mark Larkento In theory the legal systems could be much better than they are but innovation doesn’t happen very often constitutionally. In fact I haven’t seen a single constitutional change for over 20+ years and when it did happen it was negligible.
I would have to say I do not advocate putting algorithmic law into physical space because it doesn’t really fit well into that model. The US Constitution is an analog document written on paper while digital space involves smart contracts written in code. You can create smart contracts which can evolve to govern digital space or virtual space but the same processes don’t work so well outside of digital space. The reason being is that digital space has to be governed algorithmically but physical space already has governments in place.
Digital/virtual space needs order because it is the “wild west” and only the techies, hackers, programmers, can bring order to that space. Analogue space has an order already and even if we don’t agree with how it’s set up there are many stakeholders in the old order. This means there would be resistance to new technology even if it could improve efficiency simply because it might take control away from certain gatekeepers of authority and power.
Conclusion is that your idea asks for a technocracy. In the case of US law I don’t think it would do us any good if our laws were evolutionary because voters are low information uninformed decision makers, and lawmakers seek to obey the voters. I would have to say I’m against this idea because it would empower uninformed decisions and potentially amplify the ignorance.
But in virtual space or digital space where the laws don’t cost people their lives and all interaction is voluntary I can support experimentation.
https://www.reddit.com/r/bitlaw
1 Like
To illustrate what could go wrong with AI amplification of US law there could be an evolving set of laws to prevent homosexual behavior and if you make it “smart” then the law could quickly become “smart” in an “evil” way.
It could begin outlawing certain patterns (transaction patterns) associated with homosexual behavior. So in my opinion under the current environment and with the current coercive irrational controlling government I do not want to amplify the evil or promote algorithmically enhanced authoritarianism.
I also don’t support the pre-crime “crime forecasting” algorithms that police use and I would say that is bad enough. The last thing we need are laws which forecast our behavior and which oppress us in evolving fashion. I think the main difference between virtual and physical space is that in virtual space it’s easy to opt-out while in physical space there is no opting out to go to another legal system.
If I don’t like the laws in physical space I’m stuck with it. If I don’t like the laws of a virtual nation I can freely leave that nation and join another. This would mean in the short term the virtual nations would have to compete for citizens while physical nation states do not.
This is why virtual nations most always make exit easy and simple. Polycentric law is perfect for virtual nations.
I agree with most of what you are saying, Dana Edwards , but I don’t see that level of purity in implementation likely.
Many of the active projects involve current real world law schools and legal scholarship.
1 Like
Mark Larkento I’m focused on the virtual/digital world. Smart contracts don’t really have anything to do with “real world law schools” because it doesn’t have anything to do with “real world law”. Smart contracts are self enforcing and everything is algorithmic.
A smart contract can interface with “real world” legal institutions by for instance allowing for an arbitrator. This way if the parties of the smart contract want to introduce traditional law such as a judge or lawyer into the mix as an arbitrator they can.
But as for the blockchain itself it is sovereign. It’s not going to be governed by anything other than the source code. It’s not the property of any particular jurisdiction and is owned by whomever can program it.
So there has to be a clear separation between that which is digital/virtual sovereign and that which isn’t. For example I’m a physical person so if I make a profit I would have to pay taxes but whatever the laws are of physical space have nothing to do with whatever goes on in virtual or digital space.
Virtual/digital space is currently the wild west until we govern it ourselves by inventing a governance protocol. It cannot be done by the US government or any nation state. Since there are so many different tribes who consider themselves to be digital natives it’s not an option to try to have one process to govern everyone which is why you need a polystate or polycentric law.
As for purity, it’s the free market mechanism which will either allow for quality or not. For example as long as everyone can exit a virtual state and no coercion is involved then everything is fine. If a virtual state or virtual entity starts using coercion then the individuals involved in the coercion can be prosecuted under their local laws.
So this means even if you’re part of a virtual state or polystate it doesn’t mean you can break your nations laws. This means the nations laws supersede certain aspects of the virtual state which only really exists to provide order to virtual space. To put it in it’s most simple terms we will have to follow US laws whether we are members of a virtual state or not. If a virtual state becomes “enemy of the USA” then we would have to exit the virtual state in favor of another.
I’m in agreement with your points, Dana.
I just doubt that, as this movement grows, the average lawyer or politician will have the background necessary to understand the potential consequences if implemented in “real world law”.
Mark Larkento that is why it shouldn’t be implemented by those who cannot understand. I think on a local level if a small neighborhood or town wants to try to implement it that is fine but I don’t see most politicians being the people who have technical expertise.
This is what Google Island or Seasteading is for. This is why a local community can decide to do something on a city level so that we can find out whether or not it works through experimentation on the local level.
The Transhumanist Party can push for stuff like this if it deems it necessary on the local level. At this period of time though I’m not focused so much on national or geo-state politics. I don’t and will likely never have much political influence and I don’t want to be a politician.
I will leave the politics to Mike Lorrey.
We debated briefly having a software to curate the law making process in another post, I think this is now going to have full attention.
In short, to have laws something has to make it, and if that something is a piece of software, someone has to make that software. And to make that software, people has to understand it. This at the end is the story of people making laws, one way or another. Its a “Decision Support System” what we are discussing, to be clear.
I would like to have a software to automatize decisions, while, at the same time, helping those people to keep track of what is going on with the laws. If we can make either, the automation really easy to understand, or educate the population on how to operate it, we have a winning situation. I don’t mean to say the current system is better than a less optimal state of the evolutionary law, but just pointing the best scenario.
Now, to have polycentric laws is to have people stretching the ropes from every side, and someday that rope is going to blast, violently. Not to say it can’t happen, just stating the facts. If we can control the information provided by the various lawmaking nodes, making that information available to the individuals before they interact with it, and we can provide a safe venture when conflicting nodes do clash, we might have a good scenario.
What I don’t like about polycentrism is the representation of a monolithic structure that doesn’t comply interaction to change the lawmaking nodes. Is not clear if its democratic in respect to each lawmaking center, or is in fact another despotic idea. Thus, polycentrism smells like feudalism.
Why don’t concentrate on creating better DSS (decision support systems) than what we currently have? And what we currently have is nonexistent. That is a problem right there, people doesn’t know where to get data on what is wrong and what is right, instead we have Tv.
Couldn’t we make it better? Many business do work on software for this, and the common folk should have it too.
Now, this bitlaw looks interesting. I just would like to see a DSS attached to it, or it would be a real danger because clever and devious people who would like to scam others can take advantage.
But progress is a progress. And if there are more sources for these tools, I would like to take a look.
Dana’s comments:
Instead of focusing on using AI to write the laws we should first be applying AI to help people follow the laws that are already written. An app to help people adapt their behaviors to the current laws on the books (which no one can even read or understand) makes more sense than to create even more laws which no one can read or understand.
In either case the law making technology is quite authoritarian and can easily be used for lawfare purpose. Please make sure that if you create a lawmaker app that it cannot be used as a weapon.
Or if it can be used as a weapon that it’s designed in such a way to discourage it’s use as a legal instrument of war/lawfare.
On decision support I fully support decentralized decision support so if you want to make an app which can help people make better more law abiding decisions I would be all for it.
Maybe an app which tracks the status of laws on the books, and which can interpret the consequences into plain english, so persons can have a sort of pocket legal advisor.
This could help people to avoid excessively bad consequences in their activism.
AI in my opinion should solve the attention scarcity problem on behalf of the people most susceptible to risk. Lawmakers might not be making the best possible laws but AI could add heaps of complexity to an already overly complex hard to maintain legal system so that even fewer people understand it.
2 Likes
I would give 10 thousand likes to that comment alone.
Serious research has been done, and one publication I now of:
http://www.amazon.co.uk/Plunder-When-Rule-Law-Illegal/dp/1405178949
Thank you very much @Mark_Larkento for copying this most interesting conversation to the Social Future Forum! 
However, care has to be taken when copying conversations from Facebook. It should be clear in every case who is writing what. So, you need to copy the name of the participant of the conversation when you copy it. Best post the name first and set everything else as quotation (5th button from the left).
On the topic of polycentric law: I think the ideal case would be that laws become contracts which you have to agree to in order to become effective. In other words: You are only subject to the sets of laws that you have subscribed to. How would such a system be best called?
Noted, Michael. I’ll edit the post.
Please see the new category “Decentralized Applications” for related commentary copied from Dana Edwards blog.