My interest in Artificial Intelligence as a child probably had something to do with reading the phenomenal science fiction works of Isaac Asimov. I always found myself easily engaged by his writing in a way few other authors have accomplished for me.
Around the late 1990s my programming skills began to mature and with the aid of the Internet I began to read casually on the topic. The dream at the time wasn't very clearly defined, but I knew it involved an English-speaking computer that could hold a meaningful conversation with a human.
I discovered A.L.I.C.E., a computer program conceived by a Dr Richard Wallace. Designed to run on a website the program is capable of participating in a reasonably simple conversation, in English, with the remote web user on a variety of topics. I was initially facinated by this thing which behaves with remarkable similarity to many people I know.
After a short time (more than a few minutes) conversation with Alice becomes forced and unnatural. Misunderstandings are numerous and it is quickly evident that the entity with which you are speaking is nothing more than a very expensive abacus.
Frustrated by this and numerous fruitless experiments at significantly expanding the human-like conversational capabilities of the system, I came to realise it's limitations.
Around the same time I was finishing high school. In my last year in my IT class I was exposed to expert systems for the first time. I quickly constructed by own forward-chaining expert system to identify electronic components (transistors, resistors, etc.) for the fun of it and to gain a better understanding of such systems.
During class we watched WarGames, a movie about a military computer that believes it is playing a game of thermo-nuclear war when in actuality it almost starts World War III. Thanks to the efforts of a young computer hacker the crisis is averted.
It was movies like this one and later, I, Robot that continued to spur my interest in AI.
My reading continued and I discovered two significantly different strands of AI research existed; strong and weak AI. Simply put, strong AI or Artificial General Intelligence as it is sometimes known refers to the kind of intelligence we as humans and other animals we consider intelligent may possess. While weak AI is the kind of intelligence that exists in a worm, a washing machine or any other mindless automaton.
There are enough mindless automatons in this world already. There is nothing particularly interesting about them anymore, they're almost everywhere. With the rapid advance of technology there are often few computing challenges that cannot be addressed with a little time and money.
The same cannot be said for strong AI, which seems to remain an illusive goal. Although I suspect it will not be so for much longer!
Eventually I stumbled upon NARS, a creation of a computer scientist by the name of Pei Wang. The Non-Axiomatic Reasoning System is a design for an inference engine that when coupled with an appropriate control program could be the basis for a 'thinking' computer, possibly capable of solving the same kinds of problems we humans are great at while conventional computers are lost or incapable.
In 2009 I started my Masters of Computer Science. For better or worse, the institution at which I am currently studying has little in the way of artificial intelligence. The one subject they do offer is an under-graduate subject in disguise. Ironically this semester I'm revising expert systems, genetic algorithms and fuzzy logic that I first encountered over a decade earlier. And you would be forgiven for thinking the state of the art was little different to what was around in Alan Turing's time (at least 60 years ago.)
Perhaps in part due to my more recent experiences in the workforce, attempting to derive requirements from a finished and poorly documented system I feel I am now able to work more constructively on my own ideas. I am hopeful of a time very soon when myself and others like me will bring forth to the world a new era of intelligent machines, capable of solving problems as we do and interacting with us on a more human level. Numbered are the days of mice and keyboards, and beige boxes sitting motionless on our desks.