Predicting the future is very difficult (except for Nostradamus), however I had a vision of a possible/inevitable future just now.
Sometimes it's possible to view a starting position and imagine an ending position. The transition of how that something becomes the end position from the start position is the bit that cannot be imagined easily. For example I can view a start position today of the internet as an enormous network of computers and at the same time humans are on the cusp of creating something like artificial life (see link) in a very rudimentary form.
The end position I can envision is one where a giant computer "brain" links with the internet to create an "intelligence" with access to the global wealth of knowledge that the internet will become (by internet I imagine the collection of wired computers with their firewalls removed; not just the World Wide Web). Effectively this will become a sort of super intelligence only limited by the hardware.
I have established the start and end points. How could this transition happen? I think it's inevitable that something very like this will happen. It seems likely that the internet will continue to grow and encompass more and more information (including CCTV, archive data, live demographic data etc etc) and in perhaps 20 years from now we will see a global information network unimaginable by today's standards.
The artificial brain idea is something more difficult to predict at the moment. It is hinted at in the IBM research that there are brain-like fluctuations generated without adding any structure to the network. I think that in 5 years time they will have hooked up a running brain model which will have brain-like charactistics and by adding neuronal structure there should be some "intelligence" created. At first this intelligence would be mostly involving learning strageties and data storage, but that's a fairly good approximation for life. A machine running a brain program is only limited by hardware and judging by the latest trends in information systems (see link) we are on the cusp of seeing a dramatic shift in the capability of hardware which will make silicon look like steam power. I predict that within the next 20 years graphene will be used in top-end computers.
So I predict that within the next 20 or 30 years someone like IBM (or probably Google) will have a graphene based enormous computer running a brain program that is permanently linked to an unfathomably complex global network of computers.
This is certainly the stuff of science fiction (see Alphaville's Alpha-60 and Terminator's Skynet etc), but within my lifetime I hope we will start to see something like this emerging. There are dangers with this and I certainly hope that they are being considered. Does IBM have a philosopher and psychologist in its R&D division?
(the blog title translates as "I will calculate so that failure is impossible" and is a quote from Alphaville's Alpha-60 (see previous post))
No comments:
Post a Comment