Sunday, 19 May 2013

Artificial Humanity: A New Paradigm In AI

I've been doing some work behind the scenes on the site recently.  I was given a wonderful, and extremely useful present by someone very nice (thanks Erik) - a professional quality recording microphone.

I'd been wanting to get some audio recorded, and now I can - but I'm quite new at it, so there's a lot I need to work out.

While I'm doing that though, you might want to check out this, which is probably the first article I've ever written that might, if it works, destroy the Earth.

The article concerns something that occurred to me when I was putting together the signature piece for this site - One Song.  During the many months of research that went into that piece I was able to develop a very radical new way of understanding what the mind is, and what the human self is.

If you've read it, you'll know that this isn't just a nice little academic exercise - the point of the new perspective is that it allows people to undercut suffering, despair, anxiety and anger at source.  It also opens up a new understanding of something else that lies beyond them - the nature of human peace and joy, and a way to live life free of the ancient chains of human nature.

Now, while I was putting this together, something occurred to me.

If I'm right about what the mind is, and what the brain is fundamentally concerned with doing, what would that mean for artificial intelligence?

Would it be possible to build an AI along these lines, and literally create a self-aware machine?

Anyway, a couple of days ago I wrote a piece where I lay this out, lay out a very rough sketch of a potential architecture for what I call 'Artificial Humanity'.

It's written for the layman, and doesn't go too much into technical detail, partly to make it more accessible, and partly because I am basically Amish when it comes to computing, but I think the theory is solid enough.

The interesting thing about it is that if it works, it would effectively demonstrate that human beings do indeed work in the way that One Song describes.

As such, if anyone has a background in AI, and the time and resources to knock up some kind of working prototype, I'd be more than happy to work with you on it, on the proviso that if it works, we don't plug it in to any national defense systems (Terminator) to the internet (Lawnmower Man) or to any manufacturing facility it could use to replicate itself (The Matrix).

If there's nobody who's interested in building it, that's fine as well.

You can check it out at this link.

Hope you enjoy it!

No comments:

Post a Comment