Graphic Source: Clipart.com

Artificial Intelligence, we have all heard this term, but do we really understand what it means. We could look in a dictionary and if we look at Webster's online dictionary we would find the definition listed as:
1 : a branch of computer science dealing with the simulation of intelligent behavior in computers
2 : the capability of a machine to imitate intelligent human behavior
Does this really cut to the heart of artificial intelligence? For example there is no mention in the definition, of a computer getting to the point that it might actually be alive. Is this even possible? This is the great debate that has been raging for years in the AI community. How will we ever be able to distinguish if a computer is actually sentient or just following clever program instructions? On the surface this seems like an easy task, but it is one of the hardest ones ever encountered. I have talked about this before and I have come to the conclusion that if a machine has AI and is programmed correctly, it will reach a point someday where you will be able to have discussions with it and it will be able to make intelligent decisions, but will this mean that it is alive? It will seem so to us, the people interacting with it, but does that make it so?

Looking around, I found some interesting AI experiments that are taking place now. The University of Edinburgh has been doing some interesting things with computers and robots. We have to classify robots and computers together because what is a robot really? It is a mobile computing device, regardless of what shape it takes, even one that resembles a human. A new procedure has been theorized that could allow a robot to repair itself using a system that was biologically inspired. These systems mimic genetic systems. Edinburgh U. is also in receipt of a paper by two researchers that believe it is possible for not only speech recognition, which we already have in a limited way, but also that it could be tied in to plan recognition. Wow, if this is possible than a machine could understand what it reads. Some of the things that the university is looking into in this field are cognitive representation, reasoning and applying this to constraint based reasoning with respect to learning. Also dialogue structure and AI applied to music education among other things. This university is a hot bed of robot research.

Computer
Photo Source: NASA

Artificial Intelligence is being worked on all over the world. For example Jozef Stefan Institute, in Slovenia has a few departments devoted to it. Machine learning, decision support and language technologies along with knowledge management are very popular subjects there. One of the projects they are working on is a search engine that can automatically analyze free text. Will we have a search engine that can think before it gives us the search results? I am not sure if this is the ultimate goal, but if it is, I am sure it will be quite helpful. As you can see, AI doesn't have to be only in machines, it may be in software someday, but may not be able to reach the extent of intelligence that it would in special, neural computers.

Teaching logic to a computer is a lot different than teaching it to a human. You have to remember that computers carry out program instructions and right now they can not make any distinctions that their programs don't allow them to. In a way, that is what makes us humans. We have a biological computer controlling us, our brains, that has a lot more latitude than any machine made, but will this always be the case? It has been said that if computing power keeps increasing at the rate it is now, such as doubling every eighteen months, that computers will be more powerful than our brains by 2040. Yet even this has been challenged. It is now said the our brains are far more powerful than first thought and if you took every computer made since the beginning of the computer revolution, one human brain could out power them all. The human brain is now said to be so far more powerful than a computer and the difference is so great that it is almost impossible to calculate. It has a capacity of 10+8432 zeros bytes. According to an article I read, the human brain has about 100 billion neurons and some scientists were equating this with computer chip transistors and that is where they got the idea that computers will be more powerful soon. The article goes on to explain that what they didn't figure in, was the way the connections in the brain are made, giving it far more capacity than first expected.

Will someone be able to create a brain someday and hook it up inside a computing device? In a small way we are using the power of our brains to control computers. There are now computers out there that can respond to brain commands. A use has been found for them with paralyzed people and military applications. At Brown University in Providence, Rhode Island, they want to design a computer that will have human like cognitive abilities. One of the things that they are aiming for is better communication between a human and his or her computer. They not only want a computer to be able to load an image, but to understand what the image is. For example if your computer could understand images and you downloaded a map, it might be able to point out an area on that map based on what was in its memory. It would know a horse from a pig and a pig from a building. These are the first steps that a child learns when it begins to see pictures. As far as using a brain in a computer, I have to say that we are trying to do this already. I bet that answer surprised you? In Switzerland there is a supercomputer named the "Blue Brain". It is run on a network of artificial brain cells that are meant to imitate a human brain, cell for cell. The machine is in the Technical University, in Lausanne. They are hoping that they will be able to build a computer that will be able to replicate the human brain. Since it is not known how consciousness develops we won't know yet if the machine will ever become truly aware, but I guess there is a chance. The scientists hope to reach the equivalent of a rat brain in two years and finally a human brain by 2015.

Robot
Photo Source: US Navy

Will we have machines in the future that will understand us and maybe even be our friends? Knowing how things go, I guess a better question would be are we going to build machines that understand and then send them out to war? Will machines ever decide on their own that they don't like us and want to rid themselves of the crazy humans, like in Terminator? Are we really comfortable with machines that are as intelligent or even more intelligent than ourselves? I have to say that I wouldn't want a machine that had the freedom to think on its own and the mobility to do or go where it wants? If we develop cognitive machines that have brains like ours, would they be alive? Even if we suspected this, would it be right to keep them in a closet until needed or shut off? Maybe they would grow to resent this sporadic life? Shades of science fiction. Hey I just don't want to read that some person was sucked to death by their unhappy vacuum cleaner, is that too much to ask?