Thinking, dreaming, that''s all a bit too much to ask at the moment...

C''mon, even a baby can react to it''s enviroment, and recognise people. Cavemen could look towards the woods, and see a sabretooth-tiger ready to jump them. But a computer can''t seethings yet... It can''t even recognise a road accuratly enough to move over it

quote:

---

Original post by Ronin_54
Thinking, dreaming, that''s all a bit too much to ask at the moment...

C''mon, even a baby can react to it''s enviroment, and recognise people. Cavemen could look towards the woods, and see a sabretooth-tiger ready to jump them. But a computer can''t seethings yet... It can''t even recognise a road accuratly enough to move over it

---

Not at all true Ronin_54. MIT''s AI lab has a great little robot called Kismet (actually they have a variety of robots) that interacts with humans through the exchange of facial expressions and auditory signals. Kismet was designed to investigate the infant-caretaker relationship and how communication by visual and auditory cues enables the development of a two way bond between infant and caretaker. Kismet is quite remarkable to watch and can certainly recognise people that it has already met and formed a relationship with. It is able to develop and evolve that relationship with further interaction so that it can meet new people and interact with them.

On the issue of computers recognising roads... well, for over a decade there have been automated vehicle projects going around the world. While many do simplify the problem by embedding tracking targets in the road, the group at UC Berkeley developed the BATmobile almost 10 years ago. This automated system does not rely on tracking the roadway via embedded objects, but rather models the road and other drivers via visual cues (just as we humans do). It is an excellent system!

One point I would like to make is that people should make sure they know what they are talking about before saying that such-and-such is not possible yet, or hasn''t been done, be it in the field of AI or any other field. Many of the things that have been discussed in this thread HAVE been achieved - if only to a limited degree - by AI researchers around the world.

Regards,

Timkin

We have now seen many various views on this topic, so now to go a little off-track How is people''s view on connecting human brain tissue to a computer? How would this help us accomplish understanding the human brain more, and would this be some idea in the art of Artificial Intelligence?

Kenneth Wilhelmsen
Try my little 3D Engine project, and mail me the FPS. WEngine
--------------------------
He who joyfully marches to music in rank and file has already earned my contempt. He has
been given a large brain by mistake, since for him the spinal cord would fully suffice. This
disgrace to civilization should be done away with at once. Heroism at command, senseless
brutality, deplorable love-of-country stance, how violently I hate all this, how despicable and ignoble war is; I would rather be torn to shreds than be a part of so base an action! It is my conviction that killing under the cloak of war is nothing but an act of murder

Such an interface would help in understanding the mechanics of the human brain. But I''m not sure if it is possible grab enough information to be of much use. If the brain is used in this fashion with the goal of creating an AI, then I''m not sure you could even call it an AI anymore. There has got to be some kind of line that an AI ceases to be artificial anymore if this system is used. Maybe the loss of identity ?

On a side note, I''m curious of the AI advances concerning the nature of human memory. Human memory is a very complicated system. We have recent memories which are readily available. Buried memories that are hard to fetch. Life experiences which determine how strong the memory is when created. Strong memories tend to create more of an impact on larger parts of the brain.

How this system works must come into the intelligence equation in some way. No ?

Guy