Amazing Hollywood writers are nervous about artificial intelligence, aka AI, and I don’t feel so good about it myself.
When I see all these bright, intelligent and above all young talents on the picket lines, I can’t help but wonder: could we be the next columnists?
The Writers Guild of America’s strike, the first since its 2007-08 walkout that lasted 100 days, has brought a new, fascinating and troubling topic to the fore: AI.
The promise of artificial intelligence has long troubled our human minds. Like putting humans on the moon, the idea of artificial intelligence has been dreamed up for centuries, sometimes in nightmare forms as varied as Mary Shelley’s “Frankenstein” or HAL the computer in “2001” Stanley Kubrick.
Now, in an age that has seen an astonishing acceleration in the production of scientific and technological innovations, we suddenly look up and see that artificial intelligence is upon us, producing new wonders every day, as well as new and troubling questions.
We are beginning to witness a new unease at the idea that these innovations may render obsolete millions of jobs held by flesh and blood humans.
In other words, could the anti-AI backlash at the WGA signal a new Luddite phase?
You may remember the original Luddites from your history classes. They were a reaction to the Industrial Revolution in the face of machines competing with human workers, a period of great social and economic change in England in the early 19th century.
The outraged workers got up and started destroying the machines. Fearing the worst, the WGA demanded in its latest contract proposal that the entertainment industry agree not to use AI to replace writers. Unsurprisingly, the industry refused, agreeing only to “annual meetings to discuss advances in technology”. This set alarm bells ringing in writers’ rooms across the industry.
Should we sound similar alarms to the opinion punditocracy?
We’ve seen these reactionary movements rise with varying degrees of ferocity after numerous innovations in a wide variety of industries.
Perhaps now, when not only human muscles, but also the human brain, are in a new competition, could it be artists and writers who take their turn as leaders of a reaction?
Since I, too, as a career scribe, have reason to feel competition from the possibility of artificial intelligence competing with whatever intelligence tries to bring to the columnist trade, I have delved, albeit cautiously, into the world of AI that is now available on the internet.
Intrigued, I auditioned the pioneering AI called ChatGPT with a typical expert question:
“Should I trust Donald Trump?”
Without asking me, the chatbot instantly responded with an almost scolding explainer:
“As an AI language model, I have no personal opinions or beliefs,” he said. “Trust is a subjective judgment that varies from person to person. The level of trust you have in Donald Trump, or any individual, depends on your own assessment of his character, actions, and track record.
“It is important to note,” he continued, “that public figures, including politicians, can be seen differently by different people. Donald Trump, as the former president of the United States, has supporters and critics. His policies, statements and behavior during his time in office have been the subject of much debate and controversy.
“When assessing a person’s trustworthiness, it is often helpful to consider multiple perspectives, examine their actions and statements, and review objective information and facts. This allows you to form a more informed opinion based on your own values and priorities “.
Well, thanks a lot, chatbot, but your equivocation on the one hand, on the other hand, would be dismissed by most readers as hopelessly desperate.
Maybe we should talk about it the way journalists traditionally do, with your favorite drink, no matter how high-octane it is.
Clarence Page is a columnist for the Chicago Tribune.
Today’s news and more delivered to your inbox