Friday 23 September 2016

Are we going to be ruled by computers, or are we going to BE computers?

It is a peculiar thing about robots and fake humans of all kinds, but the motives of those involved in developing them can be deeply paradoxical.

Like Mary Shelley dreaming up Frankenstein as an act of revolutionary imagination, those involved in the debate about artificial intelligence are sometimes motivated by a need to remake humanity along more rational lines.

Or perhaps not so rational. Alan Turing, whose Turing Test I described in the Guardian today, was motivated partly by the incomprehension of those around him about the way he behaved.

And I don’t primarily mean his sense of logic, wearing his gasmask while cycling to avoid hayfever. I mean his gay lifestyle, at a time when homosexuality was illegal and frowned upon. See my short biography of him, Alan Turing: Unlocking the Enigma.

Perhaps it is no coincidence that Hugh Loebner, the delightful inventor who funds the Loebner Prize – his annual version of the Turing Test at Bletchley Park – is well-known for his advocacy of the right to pay for sex.

You can’t help wondering whether in these cases there is a dream of a different, more logical approach to human life, despite contemporary mores – whatever they happen to be. Computers can be programmed differently to our conditioning. But it is paradoxical – logical about passion...

The same goes in some ways for the feminist approach by Donna Haraway and others (see her Manifesto for Cyborgs), which imagines a new kind of human life, part human, part machine, which could sweep away those troublesome distinctions between male and female, straight and gay, human and animal.

This is not to dismiss artificial intelligence. I get suspicious of AI when the corporate giants use it. This kind of dreaming is the very roots of utopian radicalism, and always has been.

But I have been fascinated by the delusions of AI as well as its dreaming. Its advocates don’t always grasp that it is the sheer imperfections of human beings that make them human. Or beautiful or attractive or thrilling to sleep with.

So, no, Ray Kurzweil and others, virtual sex will never be “better than the real thing”, because the real thing revels in human imperfection and diversity. There is in fact no such thing as ‘hyper-real’. It is a concept without content from California.

It is the opposite of this – the Japanese concept of wabi-sabi, worn, imperfect, human – on which I based my thinking about authenticity (see my short book of essays, The Age to Come).

One of the comments below the line for my Guardian article suggests that AI may not happen, but the cyborg idea might. We may not be replaced by computers, but we might find ourselves increasingly computer enhanced.

The boundaries between computers and humans would begin to blur. I don’t particularly welcome this – and there are worrying implications – but I think they may be outweighed by the advantages.

In any case, it may just be inevitable.

As long as we keep a close eye on what is real and what is not. We have to remember that human beings and the virtual versions, the numerical approximations of humans, are completely different. There is no point in making AI seem closer by limiting our idea of the sheer paradoxical diversity of what humans can do,

Hence the name of this blog...

See my book Cancelled! on the Southern Railways disaster, now on sale for £1.99 (10p goes to Railway Benefit Fund).

Subscribe to this blog on email; send me a message with the word blogsubscribe to When you want to stop, you can email me the word unsubscribe

No comments: