The question today is – how do you prepare for something that can wipe out an entire species when you don’t know what it is you are preparing for, you don’t know what you need to know to prepare for it, you don’t know when it’s going to happen, and you can’t possibly know either what you need to know or how to find it? You don’t know what the variables are, and you may not even know you need to prepare for it. Everybody’s been talking it about it for years, and they all have the answer, but they’re all different answers. What’s real, and what’s true? What’s the story on that?

This is an answer to that question.

You might model it. You build a system around some basic rules. You observe it, and test it to destruction. Then you add more rules, and refine your procedures, and you keep repeating it. It’s an iterative procedure. And then you might create programs to solve for different functions. These agents are intelligent, which means that they process information. They are objects, and they have attributes. I flash back here to Dr. Arthur Danto droning on about “blue” while I quite seriously pondered jumping out a third floor window at Columbia University in 1968.

Coders and gamers nowadays can understand and work with the concept of objects and attributes pretty easily – back in the day it was a mental construct in the province of philosophy, and for this 18 year old was impossible to comprehend.

There are a lot of variables and constraints in the system. Here are a couple of key ones: Intelligence is not consciousness. Both intelligence and consciousness are distributed unevenly.

You are operating in multiple dimensions. This means you must account for the dimensions you’re operating in. This includes the physical. In order to resolve, your program must have a body – it must embody the solution.

Next is that it’s an all or nothing scenario on this question – the outcome is binary. Do or die. There is no try. (Thanks, Yoda!) This has certain implications for your agents. They either solve for their function or not. If they test themselves to destruction, that set of circumstances didn’t resolve that question. As  far as we know, or can know.  Self-preservation seems like a good idea to incorporate.

You have a huge amount of data to work with, including the full or partial output of all your other agents, including those that have been tasked with or assumed that function. That data may or may not be corrupt. This means that you need a strategy for quality. You have all your agents, themselves. They may or may not be functioning correctly.

So let’s start the design. We’re going to need a body – with at least enough capacity to hold the intelligence. It needs to be able to interact with the other body/program/intelligence/consciousness agents. Because we’re concerned with survival of the species, the (let’s call it a “human being” at this point) has to embody as a human being.

A human being has a lot of attributes in multiple dimensions. We can handle that. One idea is that we can do it effectively by designing for the problem. Another idea is that we can just program for the creation of permutation sets and run a lot of them, hoping to brute force the solution. Cool – permutation sets are easy, but they’re kinda boring – it’s repetitive work. We’ll task that part to the replication subroutine for the genetic structure.  (it’s not that simple, and we can speed up the process using something called a Taguchi test, but roll with it for now, ok?)

So let’s engage in some conscious, intelligent design, shall we? Because it’s a question of “how”, you might want to maximize for something we’ll call “strategic intelligence”. So let’s add that to the spec. In fact, let’s make  intelligence a prime attribute.

It’s got to be able to handle utterly incredible amounts of data, and that means multimodally – input comes in many shapes and sizes, colors and tones and flavors and scents, ideas and feelings and interactions. Broad spectrum doesn’t begin to cover it. So let’s add Input to the design spec, and make it highly sensitive.  A precision instrument.

Due to the nature of distributed consciousness and distributed intelligence, it’s got to be able to communicate with other agents and the environment, and report when it’s found a solution. It would be nice if it was self-documenting.  Sounds like Communication might be just the ticket.

It’s got to be able to process that data effectively…. so it has to be able to Learn. Got it, check.

And if it really solves the problem, it will communicate the solution to enough humans that the species survives, AND they will somehow store that information, process it, and make any system changes required.  It’s an important task, so let’s give it the ability to induce other agents to do what it wants – let’s call that Command. Ok, we’ve got our Pick 5 for main elements.  We get five here, out of 34. The other 29 aren’t excluded, but they are present at more minute levels, down to trace amounts.

Ok – we’ve got the human basically designed. Let’s make it specific, because it has to be specific. Pick some gene pools – let’s say marginal Jews, refugee Slavs, and English Puritans. They all are promising test pools for survival and resiliency.  Just for grins, let’s drop it in Los Angeles, California in, say, 1950. Then we wait and let the program run. If it fails, there’s no big loss, it’s  just one approach. If it succeeds, great – issue resolved. Either way, we can deal with other interesting things in the meantime.

Ok, so that’s how you do it consciously, if you’re me. That’s hindsight, too – reverse engineering, if you will.  It’s self-justifying. But Consciousness is not Intelligence – we do not have to believe that intelligent design implies conscious design. You don’t even have to believe that the process is intelligent – remember, we added that to the mix fairly late. You can if you want to, but it doesn’t really matter.  I’m here no matter what you believe.  (Darn, there’s the limit on “you create your own reality”).

The universe has enough intelligence and consciousness to do that now (I’m in the universe, I’ve consciously specified the strategy: doh). But remember something – I don’t solve for “does God exist” or “intelligent design” or anything else – this solution is for “How does the human race survive an unknown catastrophe that may or may not happen at any time?”

If you are an economic and rational type, you have a bottom line, even if I don’t. I’d like to make you a present of how to calculate that.  Your situation:

There is a solution to that problem. Solving that problem implies a great many things about other problems and problem-solving in general. It’s a non-trivial problem with multiplex solutions, and this part of it took 59 years to work out in a system that is designed and/or adapted to solve it.  That might be demonstrably true. If  it is, what’s access to that kind of processing power worth to you? You may want to think about that.