PratGPT

I am besieged by lunacy. Surrounded with stupidity. And all thanks to AI.

It started with an email which arrived last week. ‘This message has been written in a friendly but professional style’, it began. ‘Let me know if you need any changes’. I sighed, and replied ‘did you by any chance use ChatGPT to write this?’

I felt rather sorry for the musician who had sent it, seeking career help from the music business I have been privileged to run for two decades. The rest of his missive had the now-unmistakable layout of ChatGPT, and as usual, it was far too long, po-faced, and highly repetitive. These newfangled digital tools have yet to learn the wisdom of Mark Twain’s line, ‘I’m sorry I wrote you a long letter. I didn’t have time to write a short one’. Worse, my musician friend had forgotten to remove the preamble, instead copying and pasting the entire response from his digital assistant. He was, to put it mildly, embarrassed. ‘I can’t believe I left that bit in’. I made a polite reply, and didn’t chide him for letting his synapses take a nap.

Then, two days later, another muso wrote, this time asking for a letter of recommendation for an intended move to the States. ‘Sure’, I replied, ‘just let me know what you want me to say’. And then – horrors – another long-winded response, with that same typeface and layout, informing me at interminable length of the type of content it is proper to include in such a letter. I just wanted him to explain why he was moving and how he thought I could help; I didn’t really need a two page lecture on the underlying principles of letter writing. I was sorely tempted to send him the following –

TO WHOM IT MAY CONCERN

I have known John for some years. He is a great guy and a fine musician, who deserves to be heard on a wider stage. He’s also an idiot and shouldn’t use ChatGPT when asking for favours.

The final straw came on Tuesday, with what appeared to be a nice, short email from what I can only describe as an absolute banker, thanking me for a great Teams call, saying how much he’s looking forward to following up. Only snag is, we’d not had a Teams call; it had been re-scheduled. His auto-chatbox hoodgie-ma-flip had sent the message anyway, without asking.

AI, it would appear, is better at being artificial than intelligent. Or rather, its users are, abdicating personal responsibility in the belief that, because it’s AI, it’s bound to be right, even when using it undermines their credibility. It’s the lobotomisation of speech, the garrotting of reflective thought. People turn off their brains, hoping some random remote tool coming live from an over-heated data centre swallowing billions of planet-guzzling volts will do their work for them. In the process, they forget that crucial thing called ‘thinking’. I like thinking. It’s better than the alternative.

So I was not much tempted to reach for the digits myself, writing to join what I believe to be a networking club for older people in fashionable London. (I’m pretty sure it’s that, and not a front for swingers. Hope so, anyway; the only kind of swinging I like is behind a drum kit). To be considered for the invite-only membership, I had to apply. Which means it’s not very invite-only, really, is it? This involved filling in a form which included ‘tell us about yourself’.

Eschewing ChatGPT, I decided to respond using that old-fashioned, clapped out thing called ‘my brain’. But where to start? Age? Location? Education? Inside leg measurement? Should I just put a link to that Wikipedia page about me, immediately impressing them as an absolute tosser? Or copy the approach I used on a dating App in New York half a decade ago, where my profile just said ‘English. Ironic. Adjectival’?

There was no word limit, so I could have gone on endlessly, listing my alleged accomplishments until the hopefully-not-swingers banned me in boredom and withdrew the invitation they had not yet made. When I’d finished and pressed send, waiting to be judged, I wondered if I really want to be a member of a club that might have me. And it wasn’t even Groucho’s.

But, it seems, we should none of us worry our silly little under-used heads about such matters. Because the end is coming. I read in the papers this week that tech billionaires, those people who have created so much joy in the world with their algorithms and copyright theft and extreme political opinions and obsession with being ever-richer, are all busily building underground bunkers, for the moment when their machines take over. Apparently Hawaii and New Zealand are popular locations for this act of incipient cowardice. It’s all because of ‘artificial general intelligence’. Evidently this is one step up from the ordinary, boring AI we know and already loathe. It will be literally smarter than all of us, and is coming to a life near you, rather soon. The billionaires who created it are all terrified of it. But, like my earlier musical interlocutors, instead of pausing, reflecting, and taking another direction, or even switching the damn thing off, they’ll plough heedlessly on, showing the same socially-responsible attitude for which they are now famous - by scarpering. They’ll all be underground, counting their suddenly rather pointless money, as robot-inspired wars rage around the rest of us.

Mind you; looking at the state of the world and my Inbox this week, I doubt the robots could do any worse than humans. And they might remember to take the preamble out.

On the blasted moor, mad and soaking, King Lear bellows ‘who is it that can tell me who I am?’ ‘Lear’s Shadow’, replies the Fool, helpfully, in his self-appointed role of best friend and chief underminer. We all have times in our lives when we ask that question, feeling our world slipping away, wondering if we’re on the right track, or out of our depth, or both. It’s human to lack confidence, and I understand why people might lean on technology to help overcome it, throwing their intellectual hands in the air and letting AI do some heavy mental lifting, in a case of identity heft. Bit of a shame, though, when the tool becomes the master, and we stop thinking for ourselves, in a sort of cerebral smear. All those neurons, with nowhere to go. And if you echoed Lear and put ‘who am I?’ into ChatGPT, it wouldn’t know what to say. Some things, we have to work out for ourselves.

This article has been written in a pithy but longwinded style, using fingers on a keyboard. Let me know if you’d like anything changed. And thanks for a great Teams call!