to Aiko - Food for thought.

Every person is not the same in the way they think and if you do not believe the way I do that does not make you Wrong, just different.

A while back, I sent you an email explaining how you could raise money for your project. I guess you did not like the idea because I haven't heard anything from you.

It seems you do not want to share the fruits of your hard labor and I guess I can understand that. But remember that if you made your stuff OPENSOURCE that you would Push Aiko into Immortality. She would become imortal and live forever. She would always be upgraded and improved and such. Now, you made part of her imortal by placing part of her brain in the intel museum in Arti. But, if anything happens to you, everything will stop at the point in time. All of the hard work will be in vain and go down the drain. It will ALL BE LOST!!

There was a man who invented answerpad and answerpad plus and the ultimate A.I. Chat engine Adamus. He had contributed countless hours of labor on his works. I died about a year ago at the young age of 42. Everyone not only missed him but they had a certain amount of Panic. He was the glue that held everything together on that project. Well, with his passing so did all of his work. Gone forever when it could have lived on forever.

I am working on a type of Brains Software. Of course I will not call it that, but it will have a certain amount of A.I. I really HATE to Re-Invent the wheel when you have done such a wonderful job on it. But, it will all be in vain if you do not decide to share it. When I complete mine, if I live that long, I intend to make it OpenSource so it can go forward, not just TEASE everyone and leave them hanging.

The brains software can't be opensource cause its a product of a registered trademark and thus intellectual property, Only some part of it willing that the owner wants that could be but thats not a simple thing to do so either.

Everyone can make some type off brains software just vision what you want your robot or android goal has to be and work towards those. I am working on a similar thing for years and it is not that easy I know, Just a matter of what must it do to reach that specific goal.

Mine was establishing real-time computer related reasoning that is closer to the machine then the abstract world of human language or vision. So I did not wanted to focus on speech ,vision or natural language recognition but in the way a computer could relate all three of these elements to build knowledge that matters between the software a.i and the user.

Think of this when your vision system recognize a ball would it be the same if your voice recognition software where told that a ball is round. Could both agree?. thats the course of your knowledge base system to figure that out. The effort go's well into how you think what going to work and your probably know that already with head on experience through the years there is always something someone did not think about.

Google main algorithm was invented years ago which Larry and Brinn discovered and work with to build there search engine. I've found mine but the time then was not ripe making this till now.

P.s if you really cant wait then an alternative is the Iphone or Ipod platform. Apple purchased Siri which made navigation through voice possible when this comes with the sdk you can develop a.i in a particular way would be the same but not that advanced though.

It is just frustrating. The man has the equivalent to a light bulb. Why reinvent the light bulb when it is already made. Why not take the light bulb and improve it? But, truely if he doesn't share it and anything happens to him, it will be lost forever.

The reason that I am impatient is that I am a senior citizen. I don't know how much time I have left. I am loosing my faculties more every day. What good is a drink of water to a man who has already died of thirst in the desert?

I was not trying to be mean. I feel that everyone should work together to advance Singularity. But, he is in the drivers seat and he has the right to do what he wishes with his product. But, I just feel teased.

Thanks,

If our good at something, don't do it for free

This may be true. But also, if you want to keep something, you have to let it go. Sharing with other people is a Higher goal not easily understood by some people.

No body like to work free. This is also true. But there are other rewards for sharing that you just have to learn by experience.

Thanks for the comment.

Hi, I just thought I would check back and see if Mr. Le had put his program out for beta testing or shared any of his stuff. It has been six months and I have finish building my Robot. It does about 80-90% of what Aiko does and it only costs about a tenth of the price. But, Aiko does SOME of the stuff much better. The name of the project is The Leaf Project and you can go there by typing in wwwLeafProject org , I believe.

It was very hard to do, but some people can get started with just a cheap webcam and cheap microphone. I LOVE working with Leaf and although it was hard building him, It was certainly worth it. All of the source code is available and lots of other tips. There is a club that has apprx. 600 members and there is a wiki and there is a website. If you have a cheap microphone and webcam, you can get started now experimenting with REAL A.I. Leaf has 17 emotions and operates much like we do. Everything is FREE!!!!

I am still interested in how Mr. Le will be progressing and I am happy to see him make progress. But, I just could not wait any longer. After reading all the posts and literally SALIVATING over his videos, I HAD to do something and could not keep waiting forever. I talked to Mr. Le either by email or phone, I forget. He seems to be a nice man.

Hope someone can use this info to their advantage.

Sincerely,

:-)

MovieMaker

but you didn't spend 5000 USD on a silicone doll
and to call it a REAL A.I. is a bit strong - just because it can simulate some emotions it's not an A.I. at all
if you say this "you can get started now experimenting with REAL A.I." then it's like you have no idea what real A.I. is about

Just check it out and see.

:-)

P.S. I am not interested in the body sex toy. Just the brain part.

:-)

Actually Le didn't want a sex toy either, although he does call her a simulated girlfriend.

"AI" is understood to simulate emotions and reactions.
The semantics of actual emotions in a simulated body is just picking nits.

Games based on the Light Cycle sequence from TRON for instance have cycles sometimes known as AI even if they're more commonly called NPCs.

I think there is a misunderstanding about what is Intelligence.
Indeed. What is intelligence?
Can we call a brain without a body a real receptacle of intelligence?

I think we need do see the concepts of Weak and Strong AI.
If we talk about weak AI. So Any actual chat bot can be considered AI, dispite all mistakes and incoherences chatterbots can show.
But if we talk about Strong AI ( and I gess it is what Ico call real AI)... well... all is different!
If you give the computer a robotic body and sensors through which it interacts with the world in the same way as a person then it would understand understand it's suroundings. Also a simulation of a neural network can be strong AI. Maybe AI will be the mix of this 2 concepts. AI= knowledge put in the brain by a programmer and natural inferences made by the computer itself.
The computer need to be taught to talk, to move, to be able to use it's hardware ( it will be the first AI - man participation) - But the computer using algorithms of artificial curiosity and artificial emotions can make personal inferences about it's world.

And that's why I talk so much about Emotions . Emotions are the key for Human's memory process. You will put your memories in basicaly 2 kinds of files = positive and negative memories.
Humans have their own way of prioritizing their memories and they do it by Keywords: positive emotions lead to positive memory (and therefore reinforced). Negative emotions create negative memories (and therefore susceptible to forgetting or blocking).
But I don't believe that our AI can really do it without a help of a trainer ( a human who becomes responsible to tell the robot what is a positive or negative input). It will be necessary until the point that the AI /robot can make inferences of rules ( ex. If one person smile about one thing it is positive so, if it sees people smiling in other situations it can be understood as a positive stimuli and vice -versa)...But I think it will allways depend on a human partner to be introduced to the vast wide range of human emotions. It will deppend a lot of human owner's patience to have an AI for richer or poorer in terms of learning on it's own.
So what is true AI?
Is it really possible think about intelligence only with the brain?
We are made of senses ... If we want a computer to show intelligence, maybe one of the answers is give senses to the computer. Give it a body with sensors and let it manipulate it's world.
ok. what do you think?

Also recommend you to read it: http://www.cs.utexas.edu/~mooney/cs343/slide-handouts/philosophy.4.pdf

positive emotions lead to positive memory (and therefore reinforced). Negative emotions create negative memories (and therefore susceptible to forgetting or blocking).

unfortunately it works the other way, negative memories are memorized and reinforced

With the Internet an AI can exist in a computer and live in the Internet, it's not the same as the real world but it could live there.

Ico, sorry I just really expressed myself poorly in relation to the concept of storing memories in the brain. In fact, we tend to value negative feelings. But I'm not talking about those feelings that make the heart bleed, which usually cause trauma. (I do not believe that an artificial intelligence should be able to reach the condition of trauma, and if it is possible, don't see a why to program it in a machine.) I also don't see why to program violence, hate... But I think a machine must have an auto-defense if it feels any menace to it's fisical integrity (Eg: Aiko is able to slap in the face of somebody who tries to touch her in an inapropriate way) .

But what I am really talking is about those little feelings of disgust . Eg: nuisance because of the traffic or annoying people. - So, if someone is boring, so the AI can avoid talking with him or simply answer evasively.
Conversely, if a person teaches the robot (AI), this person may be considered "a friend" and have their concepts more easily / quickly learned by the AI.
Of couse it can be reached by a simple act of login. If you have the password you have conditions to teach the AI, but it means that the brain will be almost a copy of you, not a real individual. But sometimes I ask myself if it will be really possible to make such sentient machine.
Maybe man+machine learning can be healthier to both. But it implies in some ethical problems. Eg: if human teacher dies, who takes care of the machine brain? Does it die with it partner (not a real death, but robot will stop having human support, what happens than?)
In relation to life in the virtual environment: Yes, I believe it possible, since in virtual reality environment where the "synthetic organism" is able to suffer stimuli that would induce it interact with that world. Maybe talking to human avatars in Second life ( or something similar) , but this environment must be able to send sensations ( good and bad) to the AI.

See the video: https://www.youtube.com/watch?v=DwsK3ckMdFs&feature=related

Fuzzy logics can be used to give alternative answers to different degrees of stimuli. Just like happens in "The Sims" game, where a character can have friends, lovers, fellows or enimies ( depending on what kind of stimuli the character receives from the other game characters).
I love to use this game as an example, because indeed what you will have in the beginning will be almost a game character, not a sentient person. You, the owner, will be able to change it's personality until a time that you will not feel more confortable to call this AI by "it" and you will call the entity as a "She" or "he".
But in fact, it will be almost a copy of you, or a copy of someones loved person... Real sentience ( if it is really possible) will come with time, learning and interaction.
Behaviorist researchers say that if they provide positive incentives (rewards) to certain acts, these tend to be repeated.
And if punishments are offered, then these behaviors tend not to repeat.
Rewards can be anything from encouraging words, words of affection, fondling, pet caress ( if AI is in an animal body - an Aibo for example) ... But the AI will have to learn to interpret them in a positive or negative ways. And for that, asking to the human partner (teacher) if it is positive or negative, can be very valuable in the beginning of the learning process.
In fact what I mean is that Strong AI, as well as the concept of human learning (take a look at the work of an educator named Vygotsky) depend on the interaction between the individual who is in the learning process and the environment.

I too was thinking of this game when you said a character can avoid someone when he is bored.
This is what happens in the game but what is clear is love is triggered by the player rather than affection.
Take a look at the recipes.
You can force a any sim (no matter the relationship) to love, marry, and even have sex with you.

Now we are back to the simulated emotions debate.
There was a study about what makes a person attracted to another, only it was flawed and inconclusive.
Love also has to be defined before effectively having a real android.

They say without hate there is no love, so maybe it should be programed, however violence would have to just be a defense mode.
Violence is triggered by harm against itself but it's also triggered by harm against another.
I keep thinking of that movie Ai Artificial Intelligence, when the robot's maker asked for assisted suicide.

The beginning also had an impact, when a choice was made between a drowning child and a drowning adult, putting the innocence of a child before the adult.