Future Searches Performed by AI?

Source Title:
AI not as clever as some would hope...
Story Text:

After failing to converse satisfactorily with www.intellibuddy.com, Justin Millins at New Scientist discusses why AI hasn't moved on as quickly as we may have hoped. Although...

In the next few months, after being patiently nurtured for 22 years, an artificial brain called Cyc (pronounced "psych") will be put online for the world to interact with. And it's only going to get cleverer. Opening Cyc up to the masses is expected to accelerate the rate at which it learns, giving it access to the combined knowledge of millions of people around the globe as it hoovers up new facts from web pages, webcams and data entered manually by anyone who wants to contribute.

From the details in the article it sounds as though "Cyc" could become an intelligent search engine of sorts, "hoovers up new facts from web pages". Could we be seeing the next generation of AskJeeves?

Instead of it returning a list of websites for you to trawl, it may just give you the answer.



Yup, tried the conversation. Didn't go anywhere. Normally I'd start pooring booze into my conversation partner then, but I couldn't find the appropriate slot for that.

May try again in a year or so...


Can't find a date on the website for the release of Cyc. I'll be giving it a go whenever they do finally release it. www.cyc.com

Cyc is interesting...

...but like the article says, it has been around for 22 years. That doesn't mean it won't work, but some guys from my company used to work their and they were completely underwhelmed by their approach. Now, maybe Cyc has moved away from hand generated taxonomies and structured representation, but if they haven't, those systems are very brittle when it comes to something as vast as the whole of human knowledge. Just think of all the new information that's being created everyday and you'll understand why a manual system can never, ever keep up.

On a sidenote, I'd like to ask the question, is how we're defining A.I. correct? Movies and pop culture have conditioned us to think that true A.I. is HAL 9000. True, such a system could easily pass the Turing Test, but it could also kill us. My question, then, is do we really want to define intelligence as the ability for a computer to decieve and, ultimately, harm? If not, is it time to redefine A.I.? I'd much rather have an intelligent system that understands the way we process information, and then take that understanding and look into a vast knowledge observatory and pull out useful information as we need it. Wouldn't it be cool if something was constantly out there reading and getting smarter and we could simply ask questions of it whenever we wanted. Basically, I don't care if it has any feelings, just have it answer my question and I'll be set.


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.