AI and what to do about it

Thu, 06/25/2009 - 03:29 - NLN

Of all the amazing technologies on the brink of creation, one has implications far beyond any others: the establishment of superhuman artificial intelligence, or AI.

By Ben Goertzel

The AI systems in commercial use today are what futurist Ray Kurzweil calls "narrow AI" systems, which are intelligent in particular domains but lack human-level breadth of intelligence. These impressively useful systems can schedule operations and allocate resources for the military and large corporations, spot credit card fraud, make financial predictions, play chess, land aircraft and so forth. As they develop further, they will play a large role in liberating humans from toil and creating new avenues for enjoyment to pursue science, art and other rewarding activities.

But, as valuable as it is, Narrow AI pales in comparison to the creation of "Artificial General Intelligence," or AGI systems--software programs or hardware devices, which have an ability equal to or greater than that of humans: to transfer knowledge from one domain to another; create new ideas; orient oneself in a new situation; and identify problems to solve.