How [not] to Holiday. Part 5. Writing Stories with Chat GPT

Rest assured, I shall return to my normal scholarly tone in August. But for now – I’m on holiday, and this has been quite a month in more ways than I can explain. So, humour me. I’m distracting myself!

Today, a friend’s Facebook posting alerted me to the fact that sentimental stories on Facebook are often concocted by AI and have no truth in them at all.

Now, I may already have mentioned that I loathe, with a fierce and burning passion, the horrible meme of mean and entitled ‘Karens’ – inevitably white women, of a certain age. Similarly, I hate all the vile YouTube videos about a nasty ‘Karen’, usually a white American living in an HOA (Home Owner Association). I’ve heard from various sources that they, too, are made up by AI, or acted using an AI script. When I overhear an endless series of videos narrated in the same voice, same tone, same cadences – with ‘Karen sneered’, ‘Karen snapped’, ‘Karen glared’, I feel an almost visceral response. How DARE they turn my name into a mean, misogynistic slur? I don’t find them interesting or amusing – they’re pathetic and offensive. I hate my name!

But if I complain – then I’m an entitled middle-class white woman called Karen just proving their point for them by complaining. As ‘Karens’ do. (Or am I? My understanding is that the meme originated with people of the global majority who quite rightly objected to condescending treatment by a certain type of white person. I have seriously taken steps to educate myself about racism, and I don’t believe I am racist. Nor do I think I’m ‘entitled’. If anything, I apologise too often for being in the way, or daring to ask for something. This Karen is not a ‘Karen’.)

However, to test the theory that these things are written by AI, I decided to ask Chat GPT to write me a story, to see what the result would look like. How easily can one prompt a very predictable story to fit a predetermined template? And how would I rate the ‘writing’? (This is a one-off, I hasten to add, because I realise I’m wasting resources, especially water, by playing around with a robot. Moreover,  the results are frankly boring! I won’t do it again.) My AI story is about a bigoted man called Bert. Here goes:-

“Could you please write me a realistic story about an American called Bert who lived in a housing complex and was mean and intolerant to anyone who looked different. The story must end with Bert learning his lesson the hard way.”

And Chat GPT replied, Certainly โ€” hereโ€™s a realistic short story about Bert, with a strong narrative arc and a moral conclusion. (Chat GPT didn’t warn me how dull it would be!):-


Title: โ€œThe View from Apartment 3Bโ€

Bert lived in Apartment 3B of the Chestnut Grove housing complex on the east side of Cleveland. He was 67, retired from thirty years at the steel plant, and proud of never needing help from anyone โ€” especially not the government, not the church, and certainly not โ€œthose people.โ€

To Bert, โ€œthose peopleโ€ included anyone who didnโ€™t look, speak, dress, or worship like he did. He hated loud music in foreign languages. He hated headscarves and saggy pants, long beards and bright turbans, rainbow flags and prayer rugs. He especially hated when people didnโ€™t speak English around him. He never said โ€œhateโ€ out loud, of course โ€” he just called himself โ€œold-fashioned.โ€

Heโ€™d sit on the bench outside the complex with his scuffed thermos of black coffee, offering grunted nods to the other older white tenants who tolerated his gruffness. But if anyone else came by โ€” the young Nigerian nurse from 2A, the Filipino single dad from 4C, the Mexican-American delivery driver from the next building โ€” Bert’s face hardened like a storm cloud.

He muttered under his breath, sometimes louder than he thought.

One evening, someone spray-painted โ€œGo homeโ€ on the wall near the mailboxes. It wasnโ€™t signed, but everyone suspected who mightโ€™ve said something to encourage it. Bert just shrugged. โ€œWell,โ€ he said to no one in particular, โ€œmaybe they should.โ€


Then winter came.

In early January, Cleveland got hit with one of the worst ice storms in a decade. Power lines went down, roads froze over, and the Chestnut Grove complex lost heat and electricity for four days. The management company was slow to respond โ€” they’d been cutting corners for years.

Bert, stubborn and reclusive, didnโ€™t reach out for help. His old space heater didnโ€™t work without power, and he refused to go to the warming shelter set up at the local community center. “Not gonna be packed in like sardines with a bunch of strangers,” he grumbled.

By day two, the temperature inside his apartment had dropped to the low 40s. Bert bundled up in layers of flannel and tried to wait it out, but his breathing became shallow, his chest tight. On the third night, he passed out on his kitchen floor, alone.


He came to in a different kitchen โ€” warm, full of the smell of ginger tea and something simmering on the stove. A woman with dark brown skin and a soft voice was leaning over him, holding a warm compress to his forehead.

โ€œDonโ€™t try to sit up yet,โ€ she said. โ€œYou fainted from the cold. My husband helped carry you over here.โ€

It was Amina from 2A โ€” the nurse he always avoided in the hallway.

He blinked, confused. โ€œWhy…?โ€

โ€œYou live alone,โ€ she said simply. โ€œAnd we saw the light out in your window. We were worried.โ€

Beside her stood her husband, Yusuf, who offered him a bowl of soup. Bertโ€™s hands trembled as he accepted it. In the corner, their children whispered, glancing at him with worried eyes.

He stayed the night in their warm apartment, too weak to argue.


When the power came back two days later, Bert returned to 3B. He barely spoke for a week. Then, one morning, he walked down to the common area and quietly scrubbed the wall where the hateful graffiti had been. No one asked him to. He didnโ€™t say a word.

Later that month, he surprised everyone by showing up at the potluck in the rec room. He brought cornbread โ€” a little dry, but appreciated โ€” and sat with the Filipino dad from 4C. They talked about old cars. Bert even laughed once.

He never gave a grand apology. He never became warm and fuzzy. But he nodded more often. He started saying “hello” without suspicion. And one rainy evening, he left a small thank-you card outside 2A with a packet of tea tucked inside.


Bert didnโ€™t change all at once. But he learned this much the hard way:
Itโ€™s the people you push away who might just be the ones who save your life.

Now, here’s where it gets interesting. Chat GPT offered to make any changes I liked. So, let’s say that ‘Bert’ is now a New Englander called Joe, chairman of his local residents’ association, and the story takes place in a parking lot. (I don’t want to offend any Berts with a series of ‘Bert’ stories, so I’ve rung the changes.)

Chat GPT obliged. ‘Absolutely! Here’s a revised version of the story with your requested details’:


Title: โ€œLines in the Lotโ€

Joe Whitman was the chairman of the residentsโ€™ association at Maple Court Condominiums, a tidy, if slightly worn, complex in a leafy corner of suburban Massachusetts. A retired school administrator, Joe spoke in crisp, measured tones and had a fondness for pressed chinos, classic novels, and rules โ€” especially those printed in the Residentsโ€™ Handbook.

He considered himself a man of principle. Decorum, order, and “shared standards” were his watchwords. In practice, this meant keeping the hedges trimmed, the recycling correctly sorted, and โ€” most importantly โ€” the parking lot free of chaos.

The parking lot was Joe’s domain.

He had lobbied for years to have the lines repainted and finally succeeded last spring. Now, every car was expected to park precisely within the marked bays. Double parking, backing in, or โ€” heaven forbid โ€” using visitor spaces without prior notice, were cardinal sins.

Joe’s frustration grew when residents who โ€œdidnโ€™t understand the culture of the placeโ€ โ€” his euphemism of choice โ€” failed to follow the unspoken rules. A young Haitian couple had moved in recently. A Syrian family rented two units nearby. And there was the quiet Korean graduate student who sometimes worked late and came home after midnight, parking hastily across two lines.

Joe didnโ€™t shout or sneer. He simply left laminated notices on windshields, written in the same font used by the association newsletter. โ€œPlease respect community standards. Repeat infractions will be reported.โ€ He believed he was being fair.

One cold November afternoon, Joe spotted the Haitian couple โ€” Daniel and Mireille โ€” unloading bags of groceries. Their toddler was crying. The minivanโ€™s front bumper was clearly over the line, intruding into the next bay.

Joe approached, arms crossed, ready to correct.

โ€œAfternoon,โ€ he said stiffly.

โ€œAfternoon,โ€ Daniel replied, adjusting his son’s hat.

โ€œI see your vehicle is once again outside the boundary. The lines are there for a reason.โ€

Daniel looked at him for a long moment. โ€œWeโ€™ll move it in just a moment. My wife needs to carry the baby upstairs.โ€

Joe gave a tight smile. โ€œThe rules apply to everyone, Iโ€™m afraid.โ€

That evening, Mireille posted in the community WhatsApp group โ€” politely, but firmly โ€” asking for a bit more compassion from the association when families had their hands full. Joe didnโ€™t respond. He didnโ€™t like being called out, especially not in public.


The storm hit two nights later.

Freezing rain coated the lot in slick black ice. Residents stayed indoors, waiting for the salt truck. But it never came.

Around 10:30 p.m., Joe went out to inspect the situation. It was his duty, he reasoned, to ensure no one parked irresponsibly overnight. He didn’t see the ice patch beside the Honda CR-V.

He slipped.

There was a sharp crack โ€” his hip โ€” and then nothing but pain and darkness.


He woke up to voices and headlights.

Daniel and the Korean student, Ji-Hoon, had found him. Theyโ€™d been helping another neighbor move her car when they saw the shape on the ground.

They called an ambulance, covered him with their coats, and waited until help arrived. Daniel even rode along to the hospital.

Joe spent three nights recovering. The doctor said it couldโ€™ve been much worse โ€” if he’d been alone for another hour, in that cold, things mightโ€™ve turned fatal.


Back at Maple Court, Joe took a long pause from association duties. He returned slowly โ€” first attending meetings quietly, then helping organize a community clean-up day in spring.

He made no dramatic speeches. But a few changes spoke volumes.

The association adopted a โ€œcompassion clauseโ€ for parking enforcement. Laminated notices were quietly retired. Joe began greeting Daniel and Mireille with genuine warmth. He even asked Ji-Hoon about his dissertation over coffee one Saturday.


Joe never stopped valuing rules. But he learned that people mattered more โ€”
especially the ones you once only saw as problems.

Since I can’t bear my own name becoming a meme, we won’t let Bert or Joe become memes. But I couldn’t resist one more attempt! (To think I once published thirty-odd short stories and a serial in The People’s Friend, but now a robot can write sentimental stuff at the drop of a hat!) Could you tell the stories in this blog post were written by AI?  They lack any sparkle that might make a reader want to rea it again!

I asked:- Could you please write me a very short sentimental story about a Devonian man called Derek, who had a much younger wife. A bit old fashioned in his ways, he would gently grumble (sorry, ‘comment’) about anything that displeased him, and his favourite insult was to tell poor Daisy that she was becoming more difficult with age. The story must end with Derek learning a salutory lesson.

Derek, a retired railway clerk from Devon, lived in a neatly kept cottage just outside Tavistock, with his much younger wife, Daisy. He was a man of habits โ€” porridge at seven, gardening by nine, and a cup of milky tea at precisely four. Derek didnโ€™t shout or rage; no, he commented. On the weather, on the state of the world, and most especially, on Daisy.

โ€œBit extravagant with the basil again, love,โ€ heโ€™d murmur as she cooked.
โ€œBack in my day, women didnโ€™t wear jeans to the pub.โ€
And, all too often: โ€œYouโ€™re becoming more difficult with age, Daisy.โ€

Daisy, ever patient, would smile โ€” though sometimes it flickered.

One Tuesday, Derek caught a chill. Nothing serious, but enough to keep him in bed for a few days. Daisy brought him soup, fluffed his pillows, and even read aloud from The Daily Telegraph, skipping the bits she knew would make him โ€œcomment.โ€

By Friday, Derekโ€™s spirits lifted, but Daisy looked pale. โ€œJust a bit tired,โ€ she said, brushing it off. On Saturday, she fainted in the kitchen.

At the hospital, while waiting for news, Derek sat with hands trembling and heart thudding like an old train engine. A nurse appeared. โ€œSheโ€™s all right,โ€ she said gently. โ€œExhaustion. Youโ€™ll need to let her rest. And perhapsโ€”help out a little more.โ€

Derek nodded mutely. That night, back in the quiet cottage, he made tea โ€” weak, a bit tepid โ€” and placed it by her bedside. Daisy opened her eyes, surprised.

โ€œI made it myself,โ€ he said. โ€œDonโ€™t say itโ€™s awful. Iโ€™ve decided Iโ€™m the difficult one after all.โ€

She smiled, tired but amused. โ€œTook you long enough, Derek.โ€

And for once, he didnโ€™t comment. He just held her hand.


As I said, I won’t be doing this again. But I do now agree that a robot is perfectly capable of churning out sentimental, sarcastic or scurrilous material ad infinitum. They are boring stories. The second one has just the same ‘feel’ as the first – it doesn’t make you want to read on. The tone is so bland that, even in the days when doctors’ waiting rooms had magazines lying around, you’d have had to be pretty fed up to want to read to the end of these particular stories! However, it’s grammatical, and my results had acceptable language – no inappropriate academic turns of phrase.

But why would I bother? It’s not nearly as satisfying as writing about something more interesting oneself!

Weird! Discussing an Article Title with Chat GPT

Did I imagine, even 15 years ago, that I’d be discussing a potential article title with an AI function? Sure, I benefited hugely from online resources during the writing of my PhD in 2004-9. However, even social media was only just getting off the ground – I don’t remember engaging with it until after the PhD was completed. Even then, if you asked a question, it was answered by another human being.

Image by Brigitte Werner from Pixabay

However, I came up with the ‘perfect’ article title before I fell asleep last night. So perfect that I could remember it in the morning, when I checked to see if anyone else had already used the opening killer three words before. And, unfortunately, they had. Spelled the American way, the phrase had been used for an American actress’s book that was a combination of ‘lifestyle’ and autobiography. It appears that my beautiful three words described a particular kind of Southern Belle. This was absolutely not what I expected to find. Even with the British spelling, I couldn’t begin my journal article with a recent book title! Particularly when it was on such a very different topic.

I played around with a new title. Should I run it past someone else to see if it sounded appropriate? That felt a bit extreme. It’s just a title, that’s all. I know what I’m going to write about.

And then I remembered Chat GPT. I’ve experimented with it on a microscopically small scale, as you’ll recall. Really, just enough to find out for myself where its strengths and weaknesses might lie, in case anyone asked me what I thought. And then I had stepped back, wary of committing my own words to the big robot in the sky.

Today, I typed tentatively,

‘How about this for an article title:- […] ?’

Chat GPT was impressed!

Your proposed title, ” โ€ฆ ” is compelling and effectively captures the essence of your article. It suggests that the research will challenge preconceived notions or stereotypes [โ€ฆ] , making it intriguing for readers.

Well, this was encouraging. But Chat GPT wasn’t done yet!

The phrase “โ€ฆ” is a strong, attention-grabbing lead that sets the tone for a discussion of unexpected findings or reinterpretations of historical narratives. The rest of the title clearly indicates the specific focus of the study, which is both informative and precise.

Overall, it’s a well-crafted title that should resonate with your audience.

Image by Elisa from Pixabay

Am I gullible, and maybe too easily flattered? But it does make me feel that perhaps I’ve got a good enough working title to be going on with. From here on in, though, Chat GPT won’t get a look-in. I never want to end up with something that wasn’t actually written by me. I’ll check how a title sounds. I might even check out a synonym, or ask if I’m using a phrase in the expected way. But basically, I’ve done the research – now I’ll write it up.

Postscript. I also ran the title past a real, respected human being. The title stands!

Main Image by Alana Jordan from Pixabay

Crafting the Perfect Tone: Unveiling the Depths of AI Writing

In the continuing saga of my explorations into writing with AI, you’ll remember that yesterday found me getting one of my own blog posts ‘translated’ by the Jetpack bot. It didn’t have the same voice as me in the slightest. My blog posts generally consist of commentary on something I’ve been researching; discussion of my methodology; or just an update on something exciting in connection with either my role as a researcher or as a hybrid librarian/researcher. Sometimes I’ll muse upon how I feel about my own unique hybrid role.

Yesterday, the obliging Jetpack bot saw that I was excited about reaching some new milestones. I was excited in a quiet, thoughtful, academic kind of way. Jetpack-Bot, on the other hand, was breathlessly enthusiastic, inviting my readers to follow me as I celebrated these amazing achievements and exploits.

I dutifully shared Jetpack-Bot’s translation after my own blog post, so that readers could compare our written styles. At that point, I noticed a button which invited me to alter the tone of the translation. That’s what I propose to do today, with this posting.

What you have just read was all my own writing, but I adopted Jetpack-Bot’s title. Now we’ll give the bot free rein for the rest of the posting!

Here is what Jetpack-Bot created first:-

Exploring the Depths: Crafting the Perfect Tone with AI Writing

Embarking further into my AI writing experimentations, you may recall my previous encounter where I subjected one of my blog posts to the translation capabilities of the Jetpack bot. To my dismay, the resulting content lacked even a trace of my authentic voice. Typically, my blog revolves around my insightful commentary on research, methodological discussions, or enthralling updates relating to my roles as both researcher and hybrid librarian. On occasion, I also muse upon the unique nature of my hybrid profession.

During my recent venture, the ever-helpful Jetpack bot detected my excitement as I reached significant milestones. However, while my excitement resonated in a quiet, contemplative, and academic manner, Jetpack-Bot exuded an uncontrollably enthusiastic tone, urging my readers to join in on the celebration of these extraordinary achievements and ventures.

Out of duty, I shared Jetpack-Bot’s translated version alongside my original blog post, providing readers with an opportunity to compare our writing styles. It was at that moment I noticed a button beckoning me to adjust the tone of the translation. And so, today, in this new entry, I propose to seize that opportunity and sculpt the perfect tone for my content.

There’s a Choice of ‘Tone’. I opted for ‘serious’:-

(The bot seems to think that long words make for a more serious tone – what do you think?!)

Crafting the Perfect Tone: Unveiling the Depths of AI Writing

Delving deeper into my explorations of AI-assisted writing, it is worth recalling the recent incident in which I submitted one of my blog posts for translation by the Jetpack bot. Regrettably, the resulting content bore no semblance to my distinct voice. As a rule, my blog entries consist of scholarly commentary on my research endeavors, methodological discussions, or intriguing updates pertaining to my multifaceted identity as a researcher and hybrid librarian. Occasionally, I reflect upon the profound implications of my unique professional role.

Upon analyzing the Jetpack bot’s response, I discovered its unwavering enthusiasm in response to my excitement surrounding recent milestones. However, rather than mirroring my restrained and thoughtful academic enthusiasm, Jetpack-Bot appeared overtly exuberant, urging my readers to accompany me on the exhilarating journey through remarkable achievements and exploits.

Out of a sense of duty, I shared the Jetpack-Bot’s translated rendition alongside my original blog post, allowing readers to discern the stark contrast in our respective writing styles. Interestingly, I observed a button inviting me to modify the translation’s tone. Consequently, I now intend to seize this opportunity, in the present discourse, to calibrate the tone and essence of my written content with utmost precision.

Well, I think that’s enough for now, to be truthful. I wrote this in my lunchbreak, and it’s time to get back to work. I’ll hit ‘Publish’ and see what my readers make of it all.

Unveiling Milestones: Moments of Progress and Excitement

Do you keep lists, partly at least for the joy of ticking things off them? I certainly do.

In terms of milestones, today I can track significant progress: not only have I decided on the titles of my two forthcoming seminars at the University of St Andrews, but something even more momentous has occurred. My Fellowship ID card is ready before I’ve even got there. It’s all very exciting!

Moreover, the chapter I promised for someone else’s essay collection is pretty much drafted. That unidentifiable editor? Well, I sought help from a genealogical expert, and was pointed towards an individual I hadn’t really considered, in a street I already knew about. It’s a distinct likelihood rather than a certainty, but I’m happy to leave it there. So that went into the contributed chapter, and it feels a bit more ‘anchored’ as a result.

So, I now have a little time to mull it over and make sure I’m happy with it. I do enjoy writing!

(I thought I could just ask the AI bot to suggest a title, but it kindly offered to rewrite my post too. Just for the purposes of comparison, read on to see how it suggested changing the text I actually authored! Clearly, I need to be more flamboyant in my style. The bot did offer me the opportunity to change the tone. I’ll try that another time … )

NB I wrote the above. The bot suggested the title … and what follows!

Milestones: Seminars, Serendipity, and Writing Delight

Join me in celebrating the remarkable milestones I have achieved on my journey. Today, I reflect on the significant progress made as I announce the titles for my upcoming seminars at the esteemed University of St Andrews. But the excitement doesn’t end there โ€“ an even more momentous occasion awaits. My Fellowship ID card has arrived ahead of schedule, adding to the thrill of anticipation!

In addition to these accomplishments, I share the progress made on a chapter I pledged to contribute to an essay collection. Through the guidance of a genealogical expert, I uncovered an unexpected lead โ€“ an individual I hadn’t previously considered, residing on a street I was already familiar with. Though not yet confirmed, this newfound possibility has contributed depth and a sense of grounding to the chapter.

Now, with time to reflect and refine, I take pleasure in the art of writing, ensuring utmost satisfaction with my work. Join me on this exhilarating journey of milestones, serendipitous discoveries, and the joy that writing brings.

Create an imaginative title and increase your readership (hopefully!)

I was excited today to see a Twitter (okay, sorry, X) posting from WordPress about their new Jetpack facility to use AI in blogging. I’m not terribly concerned about having AI do my writing for me – I’m quite confident about my own abilities! However, when it comes to picking eyecatching titles, then I’m certainly up for it. If Jetpack can create a title which will attract readers, then why not let it try?

As a matter of policy and ethics, I shall be explicit that Jetpack’s AI has helped with the title, whilst the blog post itself is all my own work. If, at a later point, I decide to experiment further, then I’ll make that explicit too. The only input to the present posting is the suggested titles, which I have quoted below in double inverted commas. As you see, I ended up choosing my own title for this posting, since none of the suggestions were suitable. Karen 1, AI – nil.

Initial attempts haven’t enthralled me, I must say. I said I needed a title, and it suggested “How to Boost Your Productivity with These 10 Simple Tips”. That doesn’t relate to this blog post.

I tried again, asking, I need a title for this posting. “Exploring the Wonders of Nature: A Journey into the Enchanting Wilderness” doesn’t exactly cut it either.

Was I not making myself clear? Please make a title for this blog post, I asked. “Unleashing the Power Within: Empowering Yourself for Success and Fulfillment” is what Jetpack suggested.

Finally, in despair, I asked this: I am curious about using AI for creating imaginative blog post titles. Please suggest a title. Oh dear, oh dear! “Unveiling the Creative Forces: Harmonizing Human Imagination with AI Ingenuity”? That doesn’t look my cup of tea at all!

I wondered if it only created nice titles when it has been allowed to create the blog post itself, first? I hadn’t yet given the bot my full text, so maybe it couldn’t see it. Suddenly, there was the answer. (Obvious, really!)

I put all this text into a new blogpost but submitted it to AI with the final words, I need a title for this. And there it was. I haven’t used the suggested title – it was my fault for trying to write about too many things in one posting, and I confused the bot with what I was talking about!

Next time, I’ll remember to be more focused on just one topic, and we’ll see what happens…

Image by Gerd Altmann from Pixabay

AI and Chat GPT (revisiting earlier blogposts)

Today, I shared three of my own blogposts (on this site) via the Whittaker Library blog. We need to know what Chat GPT can do; where its limitations are; and what the ethical problems are around using it.

If something doesn’t feel right, we need to ask ourselves why. I never want to pass something off as my own work if it doesn’t represent my intellectual effort.