I can't put my finger on the exact day it happened, but sometime during the past decade or two I lost my interest in artificial intelligence.
I remember what it was like to be inspired by the thought of truly intelligent systems. I paid my dues playing with neural-network systems downloaded off the pre-Web net. I thought that AI research must be the most glamorous and exciting job in the software field.
Today, announcements of advances in natural-language processing, computer vision, robotics and other subdomains of AI scroll down my screen without meriting a click. The field no longer seems interesting. What happened?
Part of it, I suppose, is that AI has delivered on many of its early promises. Optical character recognition was once a pipe dream – now it's a standard feature of scanner drivers. Grammar checking relies upon techniques once the domain of bushy-haired AI researchers – now it's a baseline feature for any text-processing application. Computer games and input devices routinely incorporate algorithms that were once considered AI.
For all this progress, and despite the impressive results AI systems have achieved playing chess and Jeopardy, AI simply isn't exciting anymore. The technology has led to incremental improvements in existing applications, not paradigm-shifting world changers.
I guess I read too much science fiction as a kid. Somewhere lodged in my childhood dreams is the fantasy that one day I would speak to a computer as another being, and learn about its thoughts and feelings. Grammar-checking is too mundane a reality for the grand promise of AI. I wanted computers to come alive.
Two of my favorite books are centered around intelligent computers. Robert Heinlein's The Moon is a Harsh Mistress (http://en.wikipedia.org/wiki/The_Moon_Is_a_Harsh_Mistress), from the golden age of hard science fiction, is a recasting of the American Revolution on the moon, with an intelligent computer network named Mike playing a central role. Richard Powers's Galatea 2.2 (http://www.nytimes.com/books/98/06/21/specials/powers-galatea.html ) is a more nuanced tale of a quest to teach a rapidly evolving neural network to read and understand great works of literature. Your life will be richer if you read these books; I recommend them.
Does the thought of intelligent systems still make your heart beat a little faster? Then check out AI Jobs (http://www.aijobs.net/), a Web site that lists jobs in AI research and applications. There are still advances to be made in the field. Both the incremental advances that make the world go 'round and the paradigm-shifting ones that will banish my torpor. There are still mountains to climb in AI – go on, follow your dreams.
Web recommendation: Don't expect the links at the bottom of these musings to stick to the theme, in general. My mind usually jumps around in too undisciplined a manner to maintain such a correspondence. But this time, I need to send you to Kurzweil Accelerating Intelligence (http://www.kurzweilai.net/), a mind-expanding Web site devoted to advances in intelligent technology. The site – you can subscribe to a worthwhile newsletter there too – is part of a loose federation of Web sites and business sparked by the work of visionary Ray Kurzweil. J.D. says check it out.
J.D. Hildebrand has written hundreds of articles for dozens of publications and online communities dedicated to software development. He recently relocated to a small town outside Belgrade – stop by if your travels take you through Serbia.