ChatGPT is a fascinating proposition!

ChatGPT is a fascinating proposition!

I visualize a time when we will be to robots what dogs are to humans, and I’m rooting for the machines. — Claude Shannon

I have to say I’m on board with mathematician Shannon on this one, and who knew an arithmetic maven could actually have a sense of humor? But since we haven’t quite arrived at the era of super intelligent machines, I decided to take the infamous ChatGPT for a spin for the first time this on a cool Saturday (5/20/23) morning, and per the titular declaration, it was a fascinating proposition.

What propelled me to give it a shot was a Google newsfeed article claiming investors were turning from human advisors to our new AI friend for long-term market advice. So, I foolishly thought that Chat might be able to give me a betting leg up on today’s Cubs vs. Phillies game.

But alas, ‘twas not to be! Our AI compatriot swiftly informed me that:

As an AI language model, I don’t have access to real-time information or the ability to predict the future. My knowledge was last updated in September 2021, so I can’t provide information on the outcome of specific sporting events or any events that have occurred since then. I recommend checking reliable sports news sources or websites for the latest information on the Cubs vs. Phillies game.

Excuses, excuses! I guess I’m on my own for that one.

But that doesn’t mean ChatGPT doesn’t have some betting value. Because when I asked it how many MLB games are determined by just one run (Mline vs. Rline in bet-anese), in less that two seconds it told me it was true for 27 percent of Major League contests. My previous more detailed research indicated it was actually 28 percent, but the speed with which ChatGPT provided me with an accurate answer will prove to be magical going forward.

Then I thought, “Hmmm! For all of the obvious reasons, I’m going to do a column on suicide, and I wonder what this AI wonder could do with that?” So, I dutifully typed in “Please write a column on suicide in the style of Jeff Ward of thefirstward.net.”

I wouldn’t normally be this cordial to a machine – you should hear the invectives hurled when Alexa disappoints me – but I figured it would be propitious to be more polite to our soon-to-be artificial intelligence overlords.

And the columnar result posted below was captivating. I was pleasantly surprised that this prospect didn’t summarily end with the bot declaring, “Who the fuck is Jeff Ward and why would anyone care about what he writes?” Though, come to think of it, I’m not so sure being on the AI radar is necessarily a good thing, either.

The column was well constructed such that it approximated my stream-of-consciousness/conversational writing style more than I’d expected. Chat (we’re friends now) even managed to apply some of my mannerisms like beginning a sentence with “But,” opening the closing paragraph with “In the end,” and applying my penchant for short paragraphs.

Where ChatGPT beat me hands down was there wasn’t a single grammatical error or typo that human writers are prone to, particularly when they edit their own work. But that’s where the similarities end.

The best thing you can say about Chat’s piece, other than the speed at which it was constructed, is it’s entirely derivate. It did a great job of piecing together a series of platitudes on the topic, but it tended to run around itself in flowery repetitive concentric circles offering nothing the reader hadn’t already heard before.

I’m not knocking my new friend because AI will never experience real experience, which will doom it to an eternity of regurgitating other people’s experiences, thoughts, and insights.

It’s a lot like Gabrielle Iglesias’ or Tyler Tomlinson’s standup “comedy.” It’s well constructed and the delivery is spot on, but anyone with an IQ of over 80 can predict the next punchline with 90 percent accuracy. They have no real insights, there are no surprises, and they’re generally not that funny.

That said, Iglesias and Tomlinson have clearly found their audiences, but I bet ChatGPT could come up with an equally “amusing” routine that would eliminate the need for either one of ‘em. So, yes! ChatGPT will turn out to be the bane of, and boon for, the mediocre.

For example, it could never write a real report on an incident involving a local elected official. It couldn’t create a standup routine as intricate and multilayered as something Robin Williams or George Carlin came up with. Chat wouldn’t even be able to approximate the funny ad copy in some of First American Bank’s radio spots, because humor is a tricky proposition. It certainly couldn’t begin to approximate anything that would pass for a Ted Lasso episode.

The AI could never manufacture a book like my friend’s Sharry Lynn Blazier’s self-published This is My Elgin: Hometown Stories, Songs, and Photographs, because it never “lived” in Elgin. It could imitate her experiences, or perhaps even piece together “new” ones from snippets of hers, but unlike Sharry’s book, it would be purely fictional. The bot would have none of her insight, because it’s incapable of insight.

(The details on how to pick up a copy of Sharry’s book are included below Chat’s imitation First Ward column.)

Where it would succeed, however, is writing a newspaper article on an upcoming public event like St. Charles’ Scarecrow Festival. Chat could concoct some more than reasonable basic ad copy for a typical car dealership commercial. And it could certainly write a more cohesive, clear, and perhaps more thoughtful essay than your average high schooler or undergraduate college student, but that isn’t setting the bar too terribly high, now is it?

Which brings us to Chat’s most feared aspect, its ultimate effect on academia, which I believe is vastly overblown.

Teachers get to know their students and capabilities rather quickly. So, if someone who’s generally gotten C’s on written assignments suddenly starts writing like Nora Roberts, Houston, we have a problem. If a student who sucks at English mechanics, like I did in high school, turns in perfect work, the game is afoot.

As far a plagiarism goes, that’s been around since Grog O’Shea thought it would be funny to write dirty limericks on cave walls.

My two Filipino high school friends consistently got A’s by copying encyclopedia articles and turning them in as their own work. Perhaps that all-Asians-are-smart stereotype helped them out there. Another high school classmate regularly had his mother write his essays and he never got caught. And the possibility of buying a college or high school essay has been around ever since there’ve been colleges and high schools.

And Chat isn’t technically “plagiarizing” because, according to the definition of the word, it doesn’t copy anything verbatim.

All any self-respecting teacher has to do is hand out a couple of in-class writing assignments (no cellphones allowed) which would make it incredibly easy to detect any ChatGPT shenanigans going forward. So, I don’t think any AI will have much of a negative effect here.

Where the AI would prove to be a boon is helping a non-writer over that intimidatingly difficult getting started essay hump. It’s very difficult to teach a bad or average writer how to be a good one without rewriting their stuff. But Chat could provide great examples within seconds.

All dire AI warnings aside, and regardless of what you, I, or any of the pundits think, generative pre-trained transformers aren’t going away as long as there’s money to be made creating and improving them. In the end (There I go again!), for all the reasons we just covered, I firmly believe these bots will do far more good than harm.

 

Alright! Here’s Chat’s attempt at imitating a First Ward column. Oh! And if any of you bleeps dare make any scurrilous commentary about my writing skills, or the lack of them, I promise it won’t end well for you. I’m sure my new AI friend could come up with your home address in seconds.

Byline: A thought-provoking reflection in the style of Jeff Ward, inspired by the insights shared on thefirstward.net

In the realm of darkness that haunts our society, one issue remains persistently unresolved and alarmingly prevalent: suicide. It’s a somber topic that shatters lives, leaving behind a trail of broken hearts and unanswered questions. Today, we confront the stark reality of this tragic epidemic, peeling back the layers and challenging ourselves to delve into the depths of a subject that demands our attention.

Life’s journey can often be an arduous one, filled with tumultuous waves that crash upon the shores of our existence. Amidst these tempests, we may encounter individuals who find themselves drowning in despair, their sense of hope fading away like a whisper in the wind. These souls grapple with an internal anguish that, to them, appears insurmountable, driving them to make a choice that should never have to be made.

Yet, in our society, where the shadows of stigma often obscure the light of compassion, discussions surrounding suicide are often relegated to hushed whispers and uncomfortable silence. It’s time to break free from this oppressive cycle and confront the demons head-on.

We must first recognize that suicide is not an isolated issue but a multifaceted problem that intertwines with mental health struggles, social isolation, and a lack of accessible resources. It demands our empathy, understanding, and concerted effort to address its underlying causes.

It’s crucial to foster an environment where mental health is given the same importance as physical health, eradicating the stigma that cloaks seeking help in a shroud of shame. Only by dismantling this barrier can we ensure that those in need feel empowered to reach out and seek the support they require.

Furthermore, we must bolster our mental health infrastructure, ensuring that counseling services, helplines, and treatment options are accessible to all. No individual should find themselves standing on the precipice of despair, feeling as though there is nowhere to turn.

Beyond these systemic changes, we must also challenge ourselves to be more vigilant, more compassionate, and more aware of those around us. A simple act of kindness, an open ear, or a sincere conversation can make a world of difference to someone who feels utterly lost.

Yet, as we embark upon this journey of understanding, we must also recognize the limitations of our efforts. We cannot bear the weight of others’ struggles alone. That burden is one that requires the collective strength of a society united in purpose and compassion.

In our search for answers, we must accept that some questions may never be fully resolved. We may never fully comprehend the depths of another’s despair, nor can we predict with certainty who may be teetering on the precipice. But we can strive to create a world where no one feels as though they are standing alone on that edge, where compassion and support are readily available.

In the end, it is our collective responsibility to champion change, to fight for a society that values every life, and to dismantle the barriers that prevent those in need from seeking help. Suicide is a battle we cannot afford to lose, and together, we can forge a path toward hope, healing, and a brighter future.

Disclaimer: This column is a fictional representation in the style of Jeff Ward and is not an actual piece written by Jeff Ward or affiliated with thefirstward.net.

 

Here’s how you can get a copy of Sharry’s excellent new book:

This is My Elgin: Hometown Stories, Songs, and Photographs is a book privately published by Sharry Lynn Blazier in 2023 as a fundraiser for the Elgin Public Museum of Natural History & Anthropology. Blazier, who serves as the museum’s Director, makes no profit from this 70-page collection of funny and often poignant stories and song lyrics that consist of family and personal stories and recollections of growing up in the classic Midwestern setting of Elgin, Illinois. For information on purchasing the book, contact Elgin Public Museum at (847) 741-6655, or email Blazier at blazier_s@cityofelgin.org.

 

Leave a Reply