An interesting article about AI

[Elbonio]

http://www.wired.com/techbiz/people/magazine/16-02/ff_aimystery?currentPage=1


That link is to a gripping story of how two pioneering minds in the field of AI both committed suicide.

If you have 20 minutes to spare I would recommend it, it's fascinating, I read it over lunch this afternoon and didn't want to stop :)

[Blood Sport]

Good read.

I wonder if this is true http://en.wikipedia.org/wiki/Image:Happiness-intelligence-lisasimpson.jpg


On the subject of AI.
While i have no doubt that one day we'll create true AI, I dont think i'll see it in my lifetime

[TE-Hellfire]

While i have no doubt that one day we'll create true AI, I dont think i'll see it in my lifetime

Yes you will.

http://img99.imageshack.us/img99/2117/exponentialng2.jpg

[RocketKnight]

Nice graph, except having computer processing power doesn't mean we will therefore create true artifical intelligence. By that logic we've already created a computer capable of thought, it just thinks very slowly and we're waiting to upgrade it in 10 years time. :p: Sadly it's not quite that simple.

Also, when we do create artificial intelligence, it should be called Gizmo. :)

[Jingles]

This is assuming Moore's law holds true. Something that may break once we reach certain physical limits of the current fabrication process.

I'm expecting a plateauing followed by a sudden jump with the arrival of quantum processing.

[TE-Hellfire]

Nice graph, except having computer processing power doesn't mean we will therefore create true artifical intelligence.

You're dead right, raw computational power doesn't equate to intelligence.

However, the exciting concept is that the above exponential increase in computational power is matched within many other fields of technology, including the rate at which we are modelling cognitive structures and functions of the human brain.

For instance, just two years ago, half a mouse brain was simulated, not just matched by computational power, but an actual accurate cognitive simulation - even though it equalled 1 second of the mouse's thought processes (albeit slowed down 10 times)

If you want to know more, check out this article written by Ray Kurzweil (http://www.kurzweilai.net/articles/art0134.html?printable=1).

[andyf]

After reading a couple of his books, Kurzweil seems pretty clued-up on AI and in fact all sorts of future tech stuff. But it DOES still feel like he's very optimistic with his projected timeframes for things to happen.

Would love to be proved wrong.

[Jamz]

Are you people crazy, we all know once AI is perfected Skynet will be born and then it's game over for us!

[Neon]

Once we have created it, what will it do? What's it for? At the moment it's all about doing it to see if it can be done.

Surely this new sentient entity will be rather bored, I guess it learns everything there is to learn. Suppose this is the basis of AI led Sci-Fi novels and such :)

[Dunceantix]

Something suspicious here, I think they were actually taken out to try and stop Skynet from existing.... I've seen that documentary... what was it called... The Sarah Connor Chronicles.... ;)

[Elbonio]

To quote Thom Yorke of Radiohead:



Computers will never take over the world - there are more of us and we have hammers.

[ShedBOy]

very interesting read, although why Singh killed himself is beyond me, he had everything :/

[Lucifa]

Are you people crazy, we all know once AI is perfected Skynet will be born and then it's game over for us!

Nah, F6 quick load.