How Artificial Intelligence is plotting our Eventual Stupidity

Call me old-fashioned, but up until recently “AI” meant nothing more to me than the sound you make at the end of the day when you lay on the couch. After learning about present-day “Artificial Intelligence” — and what it’s capable of — I’ve been exhaling a more existential-like “AAYEE!”

For cave-dwellers like myself who may have not have been paying attention, OpenAI, a leading Artificial Intelligence startup, unveiled its “ChatGPT” last November. Its most recent version can effortlessly mimic human communication and create everything from world-class academic essays to Elizabethan sonnets in the voice of Winnie the Pooh. It can make (attempted) medical diagnoses, write (pretty) funny jokes, plan your next trip to Iceland, and even suggest diplomatic ways to back out of dinner invitations to your mother-in-law. (Yes, I tried.) And as charming as a Shakespearean-sonnet-spouting Pooh Bear might seem to you now, apparently “Ungoverned AI” is also now ranked by Time Magazine as item #4 in the “Top Risks of 2024”. All of which brings us to the very quotable words of Dorothy: “Toto, I have a feeling we’re no longer in Kansas.”

Now I can imagine that there are some clever linguists on the planet who will be quick to point out that this is merely the natural progression of language and that us naysayers should just all be quiet. Sure, in mastering artificial human communication we may have accidentally turned on an all-powerful, possibly sentient, hopefully-not-too- malevolent Hal 9000. But haven’t we been enlisting machines to do our work since the dawn of time? Isn’t this really just the logical conclusion of that first wheelbarrow? And speaking of Shakespearean sonnets, wasn’t he also writing these for hire already back in the 1500’s? In fact, that last greeting card you bought with all sincerity — whose voice was that in? Might we all just get down from our Luddite High Horses and make peace with the probably-peaceful AI of day? (Let me just discreetly ask here that no one downloads this essay with ChatGPT — at least not with my name on it.)

But if there are any foremost canary warnings really worth listening to, I will humbly submit that perhaps it is High School teachers. And if what we are talking about is the creation of a “Great Homework Completer in the Sky” then maybe it is the English teachers in particular whose experienced squawking needs to be heard in this thinning air of reason in the cyber realm.

As an English teacher shaped by the writing elders of the past, I think it’s the ghost of George Orwell who might first be summoned (which I’m sure ChapGPT can do quite ably). Orwell sagely reminds us that “If we cannot write well we cannot think well. And if we cannot think well others will do our thinking for us.” And before any smart-aleck techie adds “then perhaps ChapGPT can pitch in!” consider also the words of Flannery O’Connor: “I write because I don’t know what I think until I read what I say.” Or perhaps the words of Harper Lee: “It’s a self-exploratory operation that is endless. An exorcism of not necessarily his demon, but of his divine discontent.” In fact if there is a common chorus coming from the voices of the past it is that writing gives us something few other things can: Self Awareness — and isn’t this ironically what we don’t hope to see in the evolution of modern AI?

If there is something the pen-pushers of old understood well it is that slow-cooked, hard-fought, messy communication is really at the heart of our humanity. Sure AI might be a little more nimble at finding exactly the right wording for that date proposal you just emailed Suzy. But just what kind of Faustian bargain are we entering in on when we trade over our very words for the soulless applesauce of an algorithm. And can you be sure that heart-warming response from Suzy wasn’t also generated by a heartless chatbot? (All of which given new meaning to the phrase “I’ll have my people call your people”.)

And herein perhaps lies the true doomsday scenario — less about sentient machines surpassing us in the evolutionary timeline, but more about an atrophied humanity. One that doesn’t only need canned Valentine’s cards, but must sadly also use AI to even fill in the blanks! Maybe the true danger is not that the “deep learning” of AI’s growth is outpacing us, but a “deep superficiality” that results from outsourcing our very communication.

Of course I may just be over-reacting. The most recent display of AI’s prowess to go viral, I’ve learned, involved writing instructions on how to remove a peanut butter sandwich from a VCR in the voice of the Kind James Bible. (“And it came to pass that a man was troubled by his Peanut Butter sandwich for it had been placed within his VCR and he knew not how to remove it….”) So maybe we get lucky and the worst of AI is nothing more than literary parlor tricks and tech headaches for English teachers.

Or just maybe AI uses that wheelbarrow to cart off our righteous minds.

For my part I’m not surrendering my pen without a fight.