What is the Future of Artifical Intelligence

     Shane and I spend most of our days as code ninjas battling swashbuckling pirates, but we take time out every now and then to ponder the important questions of life… and I believe that AI is definately an imporant question.  At the top of the perverbial food chain (or so we think anyway), what could be more important than an intelligent machine that could unsurp our position at the top?

     AI is not a new concept, it is in fact a concept that came to the forefront in the 50’s, and it was believed even then that strong AI was possible, and would be accomplished in the not so distant future. 

As AI pioneer Herbert Simon wrote in 1965: “machines will be capable, within twenty years, of doing any work a man can do.”Their predictions were the inspiration for Stanley Kubrick and Arthur C. Clarke’s character HAL 9000, who accurately embodied what AI researchers believed they could create.

It became obvious in the 70’s that this was a gross overconfidence in the capabilities of technology.  AI was sort of put on the back burner for a time. 

As the eighties began, Japan’s fifth generation computer project revived interest in strong AI, setting out a ten year timeline that included strong AI goals like “carry on a casual conversation”.^bIn response to this and the success of expert systems, both industry and government pumped money back into the field. However, the market for AI spectacularly collapsed in the late 80s and the goals of the fifth generation computer project were never fulfilled. For the second time in 20 years, AI researchers who had predicted the imminent arrival of strong AI had been shown to be fundamentally mistaken about what they could accomplish.

Perhaps we can understand now why anyone in the field of AI reasearch will flat out refuse to make any type of prediction and even mentioning “Strong AI” or “As Smart As A Human” will probably cause ninja swords to come out of the shadows behind you.  The field is not dead, it has just become much more specilized. 

For the most part, researchers today choose to focus on specific sub-problems where they can produce verifiable results and commercial applications, such as neural nets, computer vision or data mining.^[16] Interest in direct research into strong AI tends to come from outside the field, from internet entrepreneurs (such as Jeff Hawkins) or from futurists such as Ray Kurzweil.

Most mainstream AI researchers hope that strong AI can be developed by combining the programs that solve various subproblems using an integrated agent architecture, cognitive architecture or subsumption architecture.

The human brain has roughly 100 billion neurons operating simultaneously, connected by roughly 100 trillion synapses.^[18] By comparison, a modern computer microprocessor uses only 1.7 billion transistors.[3] Although estimates of the brain’s processing power put it at around 10¹⁴ neuron updates per second,^[19] it is expected that the first unoptimized simulations of a human brain will require a computer capable of 10¹⁸ FLOPS. By comparison a general purpose CPU (circa 2006) operates at a few GFLOPS (10⁹ FLOPS). (each FLOP may require as many as 20,000 logic operations).