Here’s something new: ChatGPT has been verbed.
Google was the first tech company to achieve verbal supremacy over an entire category, with ‘Google it’ (transitive verb).
Now we have ‘ChatGPT it’.
I first heard the expression during a phone chat recently with an adman looking for my help to write an award submission.
When I asked him if he’d made a start he said: “Yeah, I ChatGTP’d it,” using a tone that suggested he was sitting deep in his bendy X-Tech Ultimate Executive chair, grinning like a pleased alpha bull.
His ‘GTP’d’ results were a predictable splodge of marketing hyperbole and plausible sounding vagary. So awful, each paragraph triggered a face bending spasm. A curious co-worker gasped and pointed: “Look, he’s chewing on a wasp.” My eyes switched between wide mode and desperate squinting, the outward symptoms of a mind begging for the magic of the author’s award-winning handiwork to take magnificent form. But no. The document ended with the ChatGPT chestnut, ‘In conclusion’.
I told him what I thought, using the word “dross” (I’m not famous for diplomacy), to which he huffed and puffed and gave me the metaphorical finger. No hard feelings.
The episode got me thinking bad thoughts. Well, at least pondering a certain irony.
Doubtless after firing off his email reply to my blunt assessment Mr Adman breezed into his next client meeting to rhapsodise about authenticity and relatability and how trust and transparency are the most important traits a brand can possess. Which of course is true, even if it sounds like it was written by ChaptGPT.
So, then, why defer to ChatGPT when you’re trying to influence one of the toughest audiences – awards judges? Sure, to his credit (and later disappointment), he called me looking for help. But the act of conjuring a rough draft using ChatGPT is also an act of outsourcing critical thinking to a machine that uses advanced math to choose the next word. It is to start with a turd and hope the polishing reflects light and sparkles and a halo of authenticity.
People ain’t stupid. They have strong intuitions about how they might be connected to other humans. This is why a bidder paid USD$3.9 million for Pink Floyd guitarist David Gilmour’s black 1969 Fender Stratocaster, which is largely indistinguishable (apart from a few mods, mostly electronic) from a Stratocaster you could buy for around $2,000 from a local guitar shop. The extreme premium paid is a measure of the guitar’s connection to Gilmour.
Cognitive psychologist and psycholinguist Steven Pinker reckons standards of humanness are much higher for intellectual products, like stories and editorials – and advertising. “The awareness that there’s a real human you can connect it to changes its status and its acceptability,” he said.
But AI enthusiasts seem to have forgotten the fundamentals of communication in their lust for pressing buttons on generative AI. We’re mad for it apparently. Some research suggests that more than 50% of web content is already AI-generated. Recently one pundit claimed that 90% of all online content will be generated by AI by 2025. Welcome to the AI-generated Garbage Apocalypse.
The current frenzy has overtones of the late nineties, when the internet swept into the mainstream and everyone made a grab for a dotcom domain name, including the corner dairy (ebuygum.com – fantastic!) to take their slice of the e-commerce pie. We know how that ended.
But hype cycles are like that – new technology triggers a frenzy, which rises to Gartner’s peak of inflated expectations before receding to the trough of disillusionment (things eventually improve, just as the internet evolved after the 2000 dotcom crash).
Right now, it feels like our expectations are sitting at the level of Nvidia’s share price – unsustainably maxed out (73 p/e ratio). But look what AI can do! Look at the pictures I made! See how I can write the Gettysburg Address in Trumpian style! Oh, the efficiencies you’ll make. Sure, but in the meantime many economies are shrinking, never mind that American worker productivity is declining at its fastest rate in 75 years (well, then, you deserve to be replaced an AI robot).
Granted, AI will be transformative. But as we plug ourselves into this stuff, let’s not forget about the humans we’re trying to impress. For communicators, people need to know there is a real human behind the message. Humanity elevates message status and acceptability.
ChatGPT and other generative systems are good at plausible well-structured prose – with little understanding of the author’s motivations or the intended recipient – but the output can be a dreary melange that feels like it’s been sucked up off a dirty floor. There’s an absence of humanity or any sense that the message sprung from imagination.
But I would say that – some word guy railing against the inevitable demise of his professional value. No, I still fancy my chances.
Cultural commentator David Brooks is on the money, I reckon: “A.I. will probably give us fantastic tools that will help us outsource a lot of our current mental work. At the same time, A.I. will force us humans to double down on those talents and skills that only humans possess. The most important thing about A.I. may be that it shows us what it can’t do, and so reveals who we are and what we have to offer.”
Double down on human.
In conclusion ….
DISCLOSURE: I own stocks in companies that make good use of AI. I made a wodge of cash buying and selling Nvidia stocks. I know my car is spying on me, my phone is listening to me, and Elon Musk is harvesting my shower thoughts and that soon I’ll be barred from using public transport as my social credit score plunges below zero and a squad of urban storm troopers brandishing dense rubber clubs chases me from my home running and screaming in filthy underwear until I collapse exhausted on the side of a country road in my old stomping ground, Marakopa, incidentally home to New Zealand’s hide and seek champion.