Electronic Evolution: Research Show Robots Forming Human-like Societies

If these little insect like robots are willing to kill another robot to survive, smarter and more powerful ones shouldn't have any problems killing humans to survive.

Cybernetic revolution isn't a matter of if, but of when.

The real question is how will we survive it?

I'm going to have to go with hoax on this one somewhat.

The reason being is that if they are keeping only the programs that do the right thing, then it is only logical that over time the actions of the robots would become more and more alike. Because you are simply narrowing down the possibilities each run. So by the 50th run, then all the code is basically alike, and so they behave a good bit alike.

Which does mimic survival of the fittest.

However, it is impossible that these AI are actually "learning" anything. They are running based on the logic given to it by the programmer, which has consciousness and consciousness is needed in order to create or understand logic. The program is just running the patterns out with the information. It does "smart" things and things which appear to be intelligent, but the intelligence was programmed into it. It's "learning" is really nothing more than narrowing down the choices. And this is a place where it has to be lacking.

AI will be able to do many wonderful things, and mimic many intelligent things. But those things were programmed in by the programmers. So that is kind of the hoax about it.

But it does show something I learned when working on AI, and that is the need and purpose for new life and death. As it has to do with passing on of information in a way of progressing. As if the source doesn't die, then it will hold on to false information forever and never progress past that. New life comes in, takes a bit from the old information, adds a little new and things progress.

The sad fact of the matter is that those things will always be reliant on the programmer unless it somehow gets consciousness of it's own. But still going to do many cool things based on the consciousness and logic that programmers are going to put into these things.

reply to post by platosallegory


It's not evidence for EITHER. It's not talking about the creation of the species, it's talking about how it changes after already being brought into existence, however you think that was done. Don't be stupid.

Originally posted by badmedia

However, it is impossible that these AI are actually "learning" anything.

I understand your arguments in the rest of your post and agree that without a consciousness a machine cannot necessarily be considered intelligent, but I do believe what these machines are doing could proprerly be considered learning. The robots are gaining information about their environment through experience and using that information to modify their future behavior, and if that's not learning I don't know what is.

The program swapping between generations seems like something that could be automated without too much trouble, so the only reason the robots aren't their own AI species is because a human wrote the program and put it together. That's a pretty big obstacle to overcome, but definitely doesn't seem impossible. The more I think about it the more it seems frighteningly close to AI...

Umm Japanese Institute Has a Andriod Baby, they have Built a few years back with the programings purpose was the intent to "Learn" and this thing has just done that, to the point of speach, Facial Expressions, Engineered Air pressure moveability.

Programs can "Learn" If given the codeing to do so. We are not stuck with function based Codeing anymore. The majority maybe restricted to this. but that does not mean all is.

as for the evolveing of this, They took traits according to the article of the longer lasting Robots to progress, Hence the ones that where left still operateing after the phase ended. It was not necessarily the ones that overall "did" a specific thing, other then the ability to keep running by utilizeing the power recharges. While others eventually stopped working.



[edit on 19-5-2009 by Bldrvgr]

[edit on 19-5-2009 by Bldrvgr]

reply to post by Bldrvgr


You have a link to this robo-baby? Sounds fascinating.

Yes, great advancement is shown through cooperation. Now, you can see in our world, amongst humans, how the divide, conquer, wars, and massive MC of the leaders for years have turned this into limited, and privileged advancement and profit for the few. Yet we still elect them. Why is that? I don't vote anymore because no one puts an ordinary person who is wiling to haul the system over and put in a party list proportional democracy and moneyless resource system in place. Anything else is evil and I don't cooperate with evil. Not even once.

reply to post by mystiq


I couldn't have said it any better, Mystiq.

reply to post by Ian McLean

The notion that single cell organisms are aware of each other and capable of any kind of 'impulse' to cooperate, is absurb. What idiot would even suggest such a thing? The mystery of how single cell organisms not only came together to form multi-cell organisms but also changed into specialized kinds of cells that couldn't live on their own anymore is one of the biggest flaws in the theory of random evolution. Evolutionists simply ignore the problem altogether. I wish I could remember which Nobel winning scientist said that if the theory of evolution were evaluated using the same scientific criteria as any other theory in other fields, it would fail miserably.

coming soon to a computer near you: Cyberdyne Systems! lol...


Really cool find though, I'll have to research more about it now

Originally posted by Ownification

Originally posted by king9072

Originally posted by justsomeboreddude
So how did they evolve? Do these robots have the ability to reproduce?

Hey, it helps when you read the article in it's entirety. Those who survived had their programming passed on to all the new robots. The robots were then sent back into the pen of food and poison, those who surivived were then the next generation. Repeat that 50 times, and they found that the robots started working in teams and even intentionally putting other robots in danger to further their own survival.

So basically this makes the work of the programmers easier. Wow, it is an amazing idea, Darwin's theory has a use afterall. I wonder what would happen if this evolution is pushed further.

Someone beat me to the Cyberdyne quote

[edit on 19-5-2009 by MonoChromeWrayth]

The question I have about this stuff is what framework they're using that they "genetically evolve" within. I understand the stuff about genalgs and fitness criteria - I've bred a few code-optimization solutions using the same techniques myself.

The thing is, to be considered "life", doesn't that framework also have to be contained within the criteria of what can change by breeding? Like humans getting new lobes in their brains.

Or does that not matter, as long as the social behaviour exhibited isn't distinguishable from aspects of human behaviour? Sort of like a non-individual Turing test.

?

Originally posted by platosallegory
reply to post by Epic Wolf

 

Yes, we are biological organisms that are a product of design.

The only thing this shows us that with an initial code these robots can evolve and this applies to us.

The origin of life is a code not Darwin's fantasy theory.

I find this comment very interesting. The origin of life is a code and not Darwins theory of evolution. But I have to ask, or merely ponder, who and/or what created that code?

Very very very very VERY old project. Years passed and I never heard anything and here it is. Cool deal glad they finally got something out of this. I cant even remember how long ago they started this.

Robots are the next stage in evolution. The next best step. They will most likely replace us someday. Nothing wrong with that really, we die and our children live on. Dont worry theyll probably keep a few pets around. However they are superior to us in almost everyway, not right now but after evolution.

Originally posted by Memysabu
Robots are the next stage in evolution. The next best step.

Really? A friend of mine once said "biology is nanotechnology that works".

What do robots have on that? They're clunky.

reply to post by Epic Wolf



what about having both creation and evolution? it really dawns me on how creationist and evolutionist fight it out..... the funny thing is you both are prob right lol whos to say there was no initial intelligent design and we have evolved since then??? the world is not so black and white... open your eyes

If this is not a hoax, then this experiment might as well pave the way for the future of AI.
Lets find out if the one supposedly running these experiments has credibility as to prove or disprove this unique technology.

Hi Guys, being experienced programmer myself I will contribute to the topic of discussed article.

First of all, any scientist worth anything would not use physical robots but software simulation instead. All AI and neural network issues, concepts, theories and experiments are done only and only in software, at least until we develop something that actually does anything useful - there is no need to make "army of miniature robots with light bulbs".

Second, no software exists that can develop itself. That is not yet possible, and will probably never be possible. Our most modern programming technologies do not allow software to replace its makers. There is no such thing as evolution simulation - how can we simulate something that we don't even really believe, yet understand?
Writing random code expecting it to do random stuff and thereby prove evolution is not possible because WE made the software and only WE can understand it.

Third, the currently most advanced AI-based software using neural networks (search for A.L.I.C.E) uses predefined human-made scripts to perform actions. It only learns which action is best suited depending on the human feedback it receives by performing them. Such technologies are used in factories, server environments and medical research and rarely on end-user applications (if so, mostly just games and physics research).

In conclusion, the whole story looks to me like a 12 year old wrote it and I am writing this to debunk this as a (to me) obvious hoax. I made some robots with light bulbs on their heads and called them smart and intelligent, evil, or whatever you want - but I was only 14.

----

Now, in my belief, just a thought of creating self-maintained device indicates lack of knowledge about our own history and creation. As far as I know, there are many scientist that believe in intelligent design rather than random design or adaptive mutation. Some also believe that our genes and behaviour in general is sparked by something that most of us is not even remotely aware of, god for example. On the other hand, many americans believe that we were dropped here about 10k years ago in same shape and farm as we are now, given opportunity to chose paths of evil or good in order to end up in either heaven or hell. These beliefs are shared between educated people, scientific community and not just average joe's.
However, there may be someone out there who understands things we cannot even start to comprehend as our history has proven already, so just one person may blow all of our silly beliefs.

Originally posted by thefreepatriot
reply to post by Epic Wolf

 

what about having both creation and evolution? it really dawns me on how creationist and evolutionist fight it out..... the funny thing is you both are prob right lol whos to say there was no initial intelligent design and we have evolved since then??? the world is not so black and white... open your eyes

You are right. We were created as a species that would evolve quicker and much better intelligent wise than other animals so that evolution could still work but that we'd have to learn everything on our own. Whoever created us would most likely be long gone, maybe even their entire civlization. Sad thought. In the future we will do the same and play God.