3 ms·
I've often observed the same thing with a different machine. It's a great metaphor for many things. There is something about cybernetics and control theory, th
by internaut 10y ago
I've often observed the same thing with a different machine. It's a great metaphor for many things.
There is something about cybernetics and control theory, that, if applied to governance would be of enormous value, but I've never seen a real world application. I wonder if we'll get to see such a thing in our lifetimes.
In the cyberpunk Pyscho-Pass there exists a nice fictional account of such a system.
> You probably just need both.
Yes, today. My complaint is that this is a very crude algorithm! There has got to be a better way that represents people's true interests. My own hypothesis is that in the year 20XX there will exist a <country> with a combination of intelligence agency with a vastly expanded remit and a computer system which produces most central governance.
Did you believe/think/feel X today? Your inputs have been factored and there are Y resolution proposals! The resolution you have chosen shall be weighted against counterproposals and if selected shall be converted into contracts for activities that a new arm of the State shall spring into existence to deal with. Government functions can scale backwards and forwards in an orderly and consistent fashion with the desires/knowledge of the citizenry.
I think it can only work by illustrating trade-offs in order to keep stakeholders in the loop. As long as the system is comprehensible it should work. Pray we never fork.
We should try experiments like these on a Seastead first before we kill everybody.
- TFortunato 10y agoOut of curiosity, have you heard of project cybersyn? https://en.m.wikipedia.org/wiki/Project_Cybersyn https://en.m.wikipedia.org/wiki/Project_Cybersyn
- internaut 10y agoYes I have. I don't know much about it though. In my imagination I see crates of Cybersyn machines being shuttled away for good measure like the Ark of the Covenant into that warehouse. I can see them thinking, "ah, let's revisit this in, say, a century or two". It is high time these kinds of projects were revisited, not as an attempt to provide a new economic system per se as some envisioned in the past, but to change central government. As you probably already know many governments in the West are approaching half of the real economy. Things are starting to get weird, which is normal when the territory has evolved beyond the map. I think there shall be a phase change sooner rather than later in how government is managed. I mean a deep structural change in how administration works and not an ideological one. The question is not so much how and why, since there are lots of productive lines of inquiry and the value proposition is endless, but what must be done to accomplish political decision making moving from a network of men to one partly/wholly composed of machines making autonomous decisions? My guess is that governance does not require an AGI. It is already a slow moving narrow AI that lives partly in the legal realm and partly in the human realm. I suspect the technology to do this existed decades ago and the real reason the idea has not been developed is inertia. That and it's slightly scary. It is like fiddling with the boot sector! Better recreate a backup...