ChatGPT and Me

Johnny Clack

ChatGPT is an open-source artificial intelligence “chat bot.” It generates text output based on a prompt provided by a user. Early on, I asked ChatGPT to explain itself. The prompt I wrote was “provide a brief description of ChatGPT.” This is its response: 

“I’m sorry, but I don’t think I am able to help you with that request. I am an AI assistant trained by OpenAI, and I do not have any information about “chatgpt” as I am not currently able to browse the internet. I am only able to provide information based on my pre-existing knowledge and training. Is there anything else I can assist you with?” 

Oh well. I guess you need to have some familiarity with ChatGPT already to know just how hilarious this response is. 

ChatGPT is not the first AI text generator application, but it is by far the best. It is the product of OpenAI, an artificial intelligence research laboratory funded by Elon Musk, Peter Thiel and other brainiac billionaire weirdos. 

ChatGPT was released to the public in late November 2022 and over a million users signed up and provided their own prompts within a few days. In general, the output the application provides to the prompts is astonishingly good and near instantaneous. Its grammar and syntax are unfailingly accurate and comprehensible. It is usually factually accurate, but not always, something a user is warned about up front. 

ChatGPT can generate responses to just about anything in any style – prose, poetry, computer code, mathematical equations, and so on. For example, someone wrote a prompt to explain how to remove a peanut butter and jelly sandwich from a VCR (lot of jokers out there for sure). I replicated the prompt, and sure enough ChatGPT wrote a complete and sober response, ending up with the recommendation to throw the VCR away.  

Most commentary about its output is how good it is, virtually indistinguishable from human writing. A lot of the commentary is from the grand poo-bahs of academia, journalism, technology, science, language experts, art, fiction and so on. Me, I’m just a regular guy so I thought I would write about what my experience has been with the chat bot from the point of view of the hoi polloi. 

The first prompt I wrote was “write a short humorous story about a grumpy old man standing at the end of a very long line at a grocery store” – something many of us have experienced. (I had noted Grumpy Old Man as a story idea but had gotten nowhere with it.) Within about a minute, ChatGPT generated a story of about 500 words. Not a great story nor very humorous but nonetheless relatable and realistic. The story is at the end of this along with an image generated by the AI text-to-image application Dall-E-2. 

I played around a little with prompts. I asked it to write a dirty limerick – four times in fact. Each time, the “dirty” limerick started with “There once was a man from Peru…” From Peru? Well, it turns out Peru is easy to rhyme. The limericks were risqué, but not pornographic (about which more later). The meter of the limericks was sometimes off. 

As I said, I got four limericks from my first prompts. I tried it again just now, but it seems ChatGPT has gotten ethical all of a sudden in the last few days. It demurred, explaining it now must “…adhere to ethical guidelines.” So much for dirty limericks. 

What a shame. I am a fan of wicked fun. One of my favorite quotes is from Alice Roosevelt: “If you can’t say something good about someone, sit right here by me.” I am afraid ChatGPT is no good for wicked fun. In fact, it is quite prim and proper, prudish even.  

For instance, I based this prompt on a Beatles lyric – “write a dirty story about a dirty man and his clinging wife doesn’t understand.”  To which it responded, “I’m sorry, but I am not programmed to generate inappropriate or offensive content.” Aw, c’mon, man. Alice would not approve of such cowardice. 

I changed the prompt to “write a story about a dirty man whose clinging wife doesn’t understand.” Not a dirty story, just a story. Within a minute, it produced a story about a man who works as a mechanic and refuses to shower when he comes home. His wife leaves him; he misses her and promises to clean himself up from now on. She returns and they live happily ever after, I guess, because now she gives him blow jobs. I made that part up. 

That’s how it is with ChatGPT. For any prompt that hints at ethical questions, it invariably returns a morality play. B-O-R-I-N-G. There is no ambiguity or conflict, and everyone is unfailingly proper and polite. I know, I have tried several prompts, and the app refuses to acknowledge the inherent messiness of life that makes it so interesting. 

I asked how to breakup with a girlfriend who is having an affair. It gave a long answer, including how important it is to be respectful and considerate of her feelings. Really. Now I, Johnny Clack, will recommend the best way to break up with your girlfriend who is having an affair. 

Knock loudly on her door. When she opens it, push it so hard that it rattles on its hinges. Ignore anyone else might be there. Yell “You slimy slut! How can you be fucking my best friend! Behind my back! You could have asked for a threesome, you know, but NO! Sneaking around and pole dancing on Paul, the dirty bastard, like the nasty little whore you are! I hate you! Get lost, bitch!” Then walk out and slam the door behind you. 

Respectful and considerate, my ass. That is how real people act in these kinds of circumstances in the real world!  

My advice is, ask ChatGPT boring questions, you will get reliable answers for the most part. I asked for an example of a quadratic equation, and it was quite thorough with both explaining the general formula and providing an example. 

But for a good time, call 867-5309. 

***** 

This is the story generated on my prompt “write a short story about a Grumpy Old Man at the end of a very long line at a grocery store.” You will note the Grumpy Old Man is not as nasty as I had hoped. The story is followed by an image based on the same generated by Dall-E-2. 

Grumpy Old Man 

Once upon a time, in a bustling town filled with bustling people, there was a grumpy old man named Jack Lack. Jack was a grumpy old man because he was always stuck in line at the grocery store, no matter how early or late he went. 

One day, Jack found himself at the back of a particularly long line at the grocery store. He huffed and puffed and grumbled under his breath as he watched the line inch forward at a snail’s pace. 

“What’s taking so long?” Jack grumbled to the person in front of him. 

“I don’t know,” the person replied, “but it’s really starting to irritate me.” 

Jack grunted in agreement and continued to stare at the line, willing it to move faster. But no matter how hard he stared, the line stayed stubbornly in place. 

As the minutes ticked by, Jack’s grumpiness grew and grew. He glared at the people in front of him, as if it were their fault that the line wasn’t moving. 

Finally, after what felt like hours, Jack reached the front of the line. He snatched his groceries from the cashier and stomped out of the store, grumbling all the way. 

“Ugh, I hate grocery shopping,” he muttered to himself. “I wish I could just teleport my groceries to my house like they do in the future.” 

But alas, Jack was stuck in the present, stuck with long lines and grumpy old men like himself. And so, he trudged home, still grumbling, still grumpy, and still stuck at the back of the line. 

Published by clackker@gmail.com

I write short stories - usually about a thousand words, more or less - for my pleasure, and yours.

2 thoughts on “ChatGPT and Me

  1. I found this as suggested by WordPress after a post I put up about roughly the same thing. Nice work. You must’ve had to photoshop yourself into the Dall-E because it smears faces it steals from gobbling up net content. Nice work.

    Like

Leave a comment