AI Allowed to Create Their Own Culture

nomoreself writes "New Scientist reports that five European research institutes are building a virtual world with about 1000 virtual (AI) citizens, in order to observe the society these artificial agents create for themselves over the course of three years. From the article: "Each agent will be capable of various simple tasks, like moving around and building simple structures, but will also have the ability to communicate and cooperate with its cohabitants. Through simple interaction, the researchers hope to watch these characters create their very own society from scratch... [further], by pointing to objects and using randomly generated "words", characters should be able to conjure up their very own language and communicate with others inside their world." One of the researchers involved thinks the dwellers of this artificial world may even develop ritualistic practices."

Comments

Learning about ourselves.

[Leroy_Brown242]

This has a lot of interesting possiblities.

One being that given a blank slate, what they tend to do on multiple runs. (Do they always end up the same place, or does chaos theory win out?)

Another, is that the AI could be programmed to have a pre-disposition, and see how they progress. (Homosexuality, self destruction, etc, etc)

And yet another could change their environments and see how they react. (Plague, overcrowding, etc, etc.)

Call me skeptical, but...

[afabbro]

...wouldn't the culture they create be a reflection of the motivations they're given? 1,000 humans are so diverse that the culture could be anything. How do you get that level of diversity into AIs, using present technology?

Seriously, I imagine even describing programmatically the motivations and desires of 1,000 humans is impossible right now. You could simplify it (Sims, most CRPGs) but then you're at my question.

I have a feeling that if they are AIs who simply need to do X, Y, and Z to survive and survival is their priority, then there will be only a sterile culture of efficiency.

This isn't my area of expertise...just musing.

(Yes, I'm aware that you could therefore say that humans are result of the motivations our creators gave us...I'm not going into that.)

I hate the asymmetry in news reporting

[exp(pi*sqrt(163))]

When someone creates a new project it gets press releases and news coverage. But when the project fizzles out because it was completely lame and didn't go anywhere in the least bit interesting nobody bother reporting it. It's like news stories are written in a dialect of C++ with constructors but no destructors.

Well I think it's time for the garbage collector of the news world. Someone who goes through year or three old science and technology magazines looking for projects where the leaders say things like "this technique will replace everything else" or "I expect my system to develop self-awareness over the next 18 months" and brings a bit of closure to them. If the project has failed then the project leaders need to be asked "what do you have to say about your extravagant claims?", "how do you feel about the grant money you frittered away?" and "how do you respond to the poeple who claimed you were a crackpot at the beginning?".

I'll have to put this story in my queue for re-examination in 2006.

Re:All I want to know...

[vertinox]

Apparently, God reads slashdot... And has moderator points. ;)

But seriously, I think the social implications of AI creating religion without aid of human intervention might be quite ground shaking.

You don't need to worry

[petrus4]

Like an earlier comment said, it took millions of years for languages to appear among humans, and that doesn't even account for whether or not the process is genuinely random, or whether there was some kind of direction involved.

I personally believe that the term "artificial intelligence" (at least as it currently applies) is misleading...Outside of science fiction, there really isn't any such thing. Even in situations where somebody's been able to come up with a genetic algorithm that produced something interesting, the AI it produced was only able to operate within its' given environment; i.e., as an expert system. Take it out of the target environment however, and it would fall flat on its face just as surely as a desktop machine after coming to the end of a shell script. There's no adaptability there whatsoever.

Computers still don't have any real capacity for dealing with novelty...the best any GA I've ever heard of has been able to do is widen the category of knowledge that a given expert system can have, and make the boundaries of said category *look* more fuzzy and organic...but in reality, it's smoke and mirrors.

Occasionally I'll see applications which stimulate my interest...the creatures in Black and White were innovative, and the Sims 2 makes reasonably good use of numerical weighting, even if the pathfinding there still sucks to a degree.

Assuming it's possible for strong AI to exist at all, (and again, I have grave doubts) everything I've seen tells me it's still anywhere between 50-200 years away. Skynet or it's equivalent won't be showing up anytime soon.

Re:Culture is for Bacteria.

[Profane MuthaFucka]

I understand introspection to not be the process of determining valid input for a system, but for determining internal states of a system. As humans go, we really don't have any clue what our own internal states are. In that way, our knowlege of our own consciousness is very similar to the knowlege of others' consciousness.

I don't think I'm a fundamentalist materialist, because if there is good evidence or good argument, I would have no choice but to change my mind. I'm just a plain old materialist.

Re:Culture is for Bacteria.

[rexstuff]

This comes down to the classic debate of dualism - wheter are minds are distinct from our bodies. I'm inclined to favour the side of the dualist - that consciousness can never arise out of a machine (or even of merely matter). To be sure, neurons are ridiculously more complex than light switches, logic gates or even more complex ICs. However, a neuron can be rather accurately modeled with perceptrons, or as a simple (analog) electronic circuit with weighted inputs. The neuron takes in some electric signal input(s) and produces some electric signal output(s). Now, to me it seems that regardless of however many of these simple (or even of more complex) tiny circuits we stack together, they will never achieve anything approaching awareness. On a different note, the article summary stated that one researcher suspected that this society may even develop ritualistic behaviour, and I am inclined to agree. If the project is performing random mutations on an agent's behaviour, then it is natural for an agent to develop a behaviour with excess process and continue to use and spread it without that agent or any other perceiving the excess process as being superfluous.

Re:Culture is for Bacteria.

[spin2cool]

If you've ever had calculus, you'll realize that at some point, putting together enough discrete intervals creates something that is for all intents and purposes, continuous.

You must also remember that a computer simulation doesn't have to work exactly the same way as the human brain. If it can accurately reproduce human characteristics to such an extent that we can't tell the difference between it and a human, then how would we know? And why would we care?

Read some Kurzweil - I recommend 'The Age of Spiritual Machines'. It'll give you something to think about.