Saviour Machine?
A Cordelia or Rosalind moment for humanity?
A turning point?
As I touched on in an earlier post, discussion of AI is seemingly everywhere and not only writers and researchers are making use of it, but also government offices and courts are starting to use it, among many others. This trend seems set to continue as the HR profession takes a look at the implications for the workplace. But there are definitely hazards, for example in using AI in workplace communication.
Meanwhile, some are pondering what this means for us as a species. Is it a turning point for us? I often think of turning points through the lens of a Shakespeare play.
The play’s the thing
In ‘King Lear’, of course, there is a key moment at the start of the drama when Cordelia refuses to play a public and high-stakes game of “who loves me most” with her ageing, raging father and two unlovable sisters. This helps to set in motion the tragic events to follow (as it’s a Shakespearean tragedy, this involves lots of bloodshed and death).
However, in ‘As You Like It’, Rosalind, similarly ill-treated by a male guardian (her “moment” is when she is banished from court), manages to turn the whole thing around, with (and a spoiler alert is probably unnecessary) lots of happy nuptials, forest romps and reconciliations to follow.
So are we having a Cordelia moment or a Rosalind one? Will it all end grimly, or will it all turn out for the best?
Chomsky on AI
No one knows, of course, but there are plenty of opinions out there.
There’s Noam Chomsky (with others) in the New York Times saying that “long-prophesied moment when mechanical minds surpass human brains not only quantitatively in terms of processing speed and memory size but also qualitatively in terms of intellectual insight, artistic creativity and every other distinctively human faculty” cannot arrive while “programs like ChatGPT continue to dominate the field of AI”.
He sounds disappointed.
Shiny new toys
Yet, there are thousands, if not millions, of boosters and early adopters, kids in the sandpit, happily trying to figure out what to do with their (our) shiny new toys. I get articles on this every day in my inbox and probably you do, too.
And there are those distinguished figures, ‘Tech Leaders’, publicly asking for a moratorium on AI development and roll-out.
Most worrying of all, for me, is when AI experts say: “we have no idea of how a deep learning system comes to its conclusions”. This is the so-called ‘black box problem’ in AI. Frankly, it’s all a bit scary.
Governments and regulators are beginning to step in (though my guess is that many are waiting for others to figure it out) and yet despite all the qualms and ethical - even existential - issues involved, the AI bandwagon rushes on, possibly hurtling towards the Singularity, which might then become our “greatest and last accomplishment”.
For a sober assessment of where things stand, I found this recent article from Prospect (registration may be required) interesting.
Earlier, I wondered whether AI was humanity’s Cordelia moment. Or perhaps its Rosalind moment. It may, of course, end up somewhere in between, with a balanced scorecard of winners and losers.
Don’t let me stay
But actually, while thinking about all this, it’s a song by David Bowie, ‘Saviour Machine’ that I’ve heard playing in my mind. I am old enough to remember David Bowie’s album ‘The Man Who Sold The World’, on which it appeared.
In the song, the Saviour Machine is the brainchild of “President Joe”. I don’t know whether Bowie’s prescience reached as far as anticipating the Biden administration, or whether he was envisaging Britain finally becoming a republic. Maybe the title and name just scanned well.
Anyway, although the machine was built, the story doesn't end well. True, the machine solves humanity’s problems, hardly something to dismiss, but then it grows dangerously bored and threatens annihilation of the world if it is not banished or dismantled:
Don't let me stay, don't let me stay
My logic says burn, so send me away
Your minds are too green, I despise all I've seen
You can't stake your lives on a Saviour Machine
Let’s all hope that the current generation of AI turns out to be a less risky kind of Saviour Machine.