Stephen:

As a student of computer science, with an interest in philosophy, I have been wondering about the
following question:

If philosophy can be regarded as the epitome of human higher reasoning capability, and yet
philosophy is not reducible to mere logic — does this imply that computers, whose "brains" rely solely
on logic-based processes, will never be able to think, or reason (or indeed, philosophize) as humans
do?

Surely AI research would be pretty much a dead-end road if this were true, if the ultimate aim is to
produce human-like intelligence (or beyond)?

===========

I think that it is true, and that the project to make artificial intelligence through what is sometimes
referred to as Good Old Fashioned Artificial Intelligence (GOFAI) is a dead end. Read Bert Dreyfus'
mid 70s book "Why Computers Can't Think" and his early 90s "Why Computers Still Can't Think" for
an excellent exposition of why.

However, there are other ways to build computers, and it does not seem to me (or Dreyfus) that, in
the final analysis, computers which use different means of processing, together with emotions and
embodiment, will never be able to think. One further requirement, it seems to me, is that computers
will need to have a 'childhood' — a developmental period when they learn to think, emote and act in a
variety of specific situations surrounded by more able 'thinkers'.

Tim Sprod