Friday, April 29, 2005





Malaysian achieves success in artificial intelligence research

"A young Malaysian PhD student has achieved a major breakthrough in artificial intelligence which can lead to the creation of 'thinking robots' in five to 10 years."

I'm skeptical. This article makes expansive claims, and I get the impression the student-inventor -- as bright as he is -- doesn't grasp the enormous leaps a computer must make before being able to "think" in the sense described.

Then again, maybe we're overdue for this. As one who thinks we will most likely need artificial intelligence to survive the next century, I hope this is for real.

4 comments:

infotheorem said...

i like how the srticle does not go into details really. The ANN is nothing really that new if it is something similar to software based neural net models. But who knows what the ANN really is or how it operates, I'm going to see if this guy has any published papers.

razorsmile said...
This post has been removed by a blog administrator.
razorsmile said...

I want this to be true so bad ... but there's way too little info; what's to like?

W.M. Bear said...

I'm pretty retro on this subject. Given the current materialist-empiricist paradigm in science(which leads to the mind-brain confusion fallacy), I think we are a long, long, long, long ways from the kind of genuine "strong AI" exhibited by, say, HAL in "2001: A Space Odyssey." And the reason for this is NOT the hardware -- I think there are a number of supercomputers now that are theoretically capable of emulating human brain function. But that's the whole problem -- people are looking at it through the wrong end of the telescope, trying to get to strong AI by mimicking BRAIN function with little or no understanding of how the MIND works and, by and large, constantly confusing mind and brain. As long as this state of affairs continues, no strong AI. And given that the paradigm is as entrenched as it seems, this state will probably continue through our lifetimes at least. Moreover, frankly, I feel that this is actually a good thing. Once machines can truly think like human beings and better, we will all be up Shit Creek, believe me (pace Ray Kurzweil). Thus, I am also very disinclined to believe the story Mac posted the link to.