Adrian Graham
ChatGPT on writing
I asked ChatGPT for some writing tips. I wasn’t sure how much of the answer was a direct copy of existing internet content, or a more sophisticated synthesis (extrapolated from different source material). Either way, the answer was almost instant, coherent, sensible, and better than most people could provide:
Start with a solid idea: A great novel starts with a strong concept or idea. Take some time to brainstorm and develop your idea before you start writing.
Create an outline: Once you have your idea, create an outline of the plot, characters, and major events in the story. This will help you stay organised and ensure that your story is cohesive.
Develop compelling characters: Characters are the heart of any story. Make sure your characters are fully realized, with their own motivations and desires.
Write every day: Consistency is key when it comes to writing. Try to write every day, even if it's just for a few minutes.
Don't edit as you write: When you're writing your first draft, resist the urge to edit. Just focus on getting your ideas down on paper. You can always go back and revise later.
Show, don't tell: One of the most important rules of writing is to "show, don't tell." This means that you should use concrete details and descriptive language to create a vivid picture in the reader's mind.
Rewrite and revise: Writing is a process, and your first draft is just the beginning. After you've finished your first draft, go back and revise, rewrite, and edit until you have a polished final product.
Get feedback: Share your work with beta readers or writing groups to get feedback and constructive criticism. This can help you identify areas where your story can be improved and make it stronger.
Be patient: Writing a novel is a long and often challenging process. Be patient with yourself and remember that it takes time and effort to create something great.
Have fun: Finally, remember that writing should be fun! Don't get too caught up in the process or the outcome. Enjoy the journey and the experience of creating something new.
Then I asked ChatGPT to write a cool opening paragraph of a novel. Its reply sounded like a dark romance story, perhaps something along the lines of a Hitchcockian psychological drama. The opening ticked a lot of boxes. I can definitely see AI being able to write an engaging novel:
The first time I saw her, she was walking along the edge of a cliff, her long hair whipping around her face in the wind. I should have been terrified for her, but instead I was captivated by her beauty and her fearlessness. Little did I know that her wild spirit would soon sweep me up in a whirlwind of adventure and danger, changing my life forever.
And what about the opening paragraph of a genre action novel? This one also ticks many of the expected trope boxes. Is it immersive, or generic? I can envisage generative AI being used to write novels. Right now, it’s more suitable for copywriting and genre fiction, but I don’t see why future iterations couldn’t handle literary fiction. Publishers could even market AI generated authors. In the mean tine, here’s its genre action opening:
The night was thick with the scent of gunpowder and sweat. The sound of shattering glass echoed through the abandoned warehouse as Detective John Smith dodged a hail of bullets. He gritted his teeth and returned fire, his aim steady despite the chaos around him. The criminal underworld had never seen a cop like him before – fearless, relentless, and utterly unstoppable. With one last burst of energy, he charged towards his target, ready to bring justice to the lawless streets of the city.
How about writing a short story? I asked ChatGPT to write a short story about an ant and a potato. The story has an Aesop’s fable vibe with a didactic ending. For something that was generated almost immediately with only three basic variables (ant, potato, story), and without any effort on my part, the result is pretty impressive:
Once upon a time, there was an ant named Alice. Alice was a hardworking ant who spent her days gathering food for her colony. One day, while Alice was out searching for food, she stumbled upon a small potato that had fallen off a nearby plant.
The potato was much larger than Alice, but she was determined to bring it back to the colony. She tried to lift the potato, but it was too heavy for her to carry on her own. Alice didn't want to give up, so she decided to ask for help.
Alice searched the surrounding area and eventually found a group of ants who were also gathering food. She explained her situation and asked if they could help her bring the potato back to the colony. The other ants agreed, and together they worked to lift the potato and carry it back to the colony.
When they arrived, the other ants were amazed at the size of the potato. It was much larger than anything they had ever seen before. The colony was able to use the potato as a source of food for several days, thanks to Alice and the other ants who helped her.
Alice learned an important lesson that day. She realised that even when faced with something that seemed too difficult to handle alone, she could always ask for help. By working together, she and the other ants were able to accomplish something that would have been impossible on their own.
From that day on, Alice made sure to always ask for help when she needed it. And the other ants in the colony learned that by working together, they could accomplish great things. The small potato may have seemed insignificant, but it taught them all a valuable lesson about the power of teamwork.
How about something more ambitious? I asked ChatGPT to write a short story in the style of Haruki Murakami. It came back with a third person story, which didn’t quite work, so I asked it to re-write the story in first person. A few seconds later this is what came back:
I never thought I would be the type of person to search for someone who didn't want to be found, but there I was, driving through the city streets in search of my ex-girlfriend, Akiko.
We had broken up two years ago, and at first, I thought I was over her. I dated other women, threw myself into my work, and even traveled to new places to try and forget her. But no matter what I did, I couldn't shake the feeling that there was something missing in my life.
One day, I heard from a mutual friend that Akiko had moved to a different city and changed her name. She didn't want to be found, but for some reason, that only made me more determined to find her.
I started by searching online, typing in every variation of her name and location that I could think of. I scoured social media, job listings, and even obituaries, hoping for any clue that might lead me to her.
It was a long shot, but one day, I found a job listing for a company in a nearby city that matched Akiko's qualifications. I drove there, heart racing, and walked into the office, trying to act casual.
As luck would have it, Akiko was there, sitting at a desk and typing away on her computer. She looked up and saw me, and for a moment, I thought she was going to run away. But instead, she just stared at me, expressionless.
We talked for a while, catching up on old times and trying to find some common ground. But it was clear that things had changed between us, that we were no longer the same people we once were.
As I drove home that night, I felt a sense of relief wash over me. I had found Akiko, but I had also realized that the person I was searching for no longer existed. I had to let go of the past and move forward, even if it meant doing so without her.
I still think about Akiko from time to time, wondering what might have been. But I know that I can't go back, that I have to keep searching for my own path in life. And maybe, just maybe, one day I'll find someone who wants to be found just as much as I do.
What about a J D Salinger style story about a teenager and a model aeroplane? I asked ChatGpt and, once again, I had to ask it to rewrite the story in first person, present tense to make it feel more like YA fiction:
I'm sitting on the grassy hilltop overlooking the valley, flying my sleek silver model airplane. I built it from scratch over the course of several weeks, tinkering with every aspect of the design until it was perfect. It has a wingspan of 24 inches, and I'm obsessed with it.
The sun beats down on my face, and I feel the warm breeze in my hair. The world seems so big, so vast, and I feel small and insignificant in comparison. But as I watch my plane glide effortlessly through the air, I know that I'm capable of great things too.
I'm consumed with thoughts of my plane - how to improve it, how to make it fly faster, higher, better. I spend countless hours tweaking and adjusting the controls, experimenting with different settings until I have it just right. And now, as I stand on the hill, I feel a sense of pride and accomplishment, knowing that I built something that can soar through the sky.
As the plane flies higher and higher, my heart races with excitement. I've never felt so alive, so free. For a moment, I forget about all of my problems, all of the things that make me feel trapped and alone. I'm just a kid, with a model airplane and the whole world in front of me.
But then, just as suddenly as the feeling came, it's gone. My thoughts turn dark, and I begin to dwell on all of the things that are wrong with my life. My parents are always fighting, and I feel like I'm stuck in the middle of their endless arguments. My little sister is always crying, and I don't know how to make her stop. And then there's school, with its endless stream of tests and assignments and social hierarchies. I feel like I'm drowning in it all.
But even as these thoughts consume me, I can't help but feel a sense of hope. Because as long as I have my plane, I know that I can escape, even if it's just for a little while. I can forget about everything else and just focus on the thrill of flying.
The sun begins to set, and I know that it's time to pack up and head home. I carefully fold up my plane and put it in its carrying case, making sure that it's safe and secure. As I walk down the hill, I can't help but feel a sense of melancholy. I know that tomorrow I'll have to go back to school, face all of my problems head-on. But for now, at least, I have my model airplane, and that's enough.
I walk the rest of the way home in silence, lost in thought. When I get back to my room, I carefully place the plane on my desk and sit down on my bed. I close my eyes and take a deep breath, trying to push all of my worries and fears out of my mind.
For a moment, I feel like I'm flying again - free and weightless, with nothing to hold me back. And then I open my eyes, and reality comes crashing back down around me.
But even as I face the challenges of my everyday life, I know that I have something special - something that sets me apart from the other kids at school. I have my model airplane, and that's enough to make me feel like I can conquer anything.
And then I asked ChatGPT what made Grahame Greene such a great writer. (I asked this question after I’d written the section (below) about him). I was intrigued to see what it would come up with. Once again its response was surprisingly good. Chat GPT’s answer reproduced many of the points I’d made, most importantly my core observation that Greene writes about moral ambiguities:
Graham Greene is widely regarded as one of the greatest writers of the 20th century. There are several factors that contribute to his success as a writer:
Mastery of language: Greene was a master of language and had an exceptional command over English. His writing was marked by a simple, elegant, and precise style that was easy to read yet profound in its meaning.
Deep insight into the human condition: Greene's writing often explored the complex nature of human beings, their struggles, motivations, and flaws. He had an uncanny ability to capture the psychological nuances of his characters and present them in a way that was both relatable and thought-provoking.
Multifaceted themes: Greene's work was not limited to any particular genre or theme. He wrote about politics, religion, love, and the human experience, among other things. His ability to blend different themes and genres into his work made him a versatile and highly respected writer.
Sense of moral ambiguity: Greene's work was often characterised by moral ambiguity, where characters were not clearly defined as good or bad, and their actions were open to interpretation. This ambiguity gave his work a unique edge and made it more intriguing to readers.
Strong narrative structure: Greene was a master of narrative structure and knew how to keep his readers engaged. His writing was marked by a careful balance of dialogue, description, and action, and he knew exactly how to build tension and suspense in his stories.
Overall, Graham Greene's great talent as a writer was his ability to create complex characters, explore deep themes, and tell engaging stories with a masterful command of language.
If you think what AI can do right now, and where it could be in 25 or 50 years time, the possibilities are breathtaking. ChatGPT’s answers provoke many questions about originality and original thought, both in machines and people. Most education is based on regurgitating existing information and extrapolating on that knowledge. How original is most thinking anyway?
As a productivity tool now, I think generative AI is likely to become a common tool, much like spell check is. It will be used as a starting point for writers to work from, or an editing tool to tweak existing text (the view point, tense, and so on). Where will the writers’ skill begin and generative AI end? This will probably be a wider cultural question, seeing that so much of what we experience in the media is already AI generated, from car adverts, to models in clothing catalogues, and product descriptions. The possibilities are almost endless.
Many people have believed that, in the future, robots might do the hard and dirty work for people (like caring work and cleaning), but, there’s another possibility… what if human labour ends up doing the less skilled, dirty jobs, the hard graft, and generative AI takes over the professional and administrative roles? What happens when media consumption is dominated by super-slick, generative AI (especially in advertising and the entertainment media)? I suspect that many people will look for something different, something analogue perhaps?
Other factors to consider, are: how will we manage AI, what kind of regulation and oversight will it require? We could be in for another era of technological disruption and social upheaval that makes the rise of internet seem like nothing in comparison.
The Hunger
The Hunger (1983) is a vampire horror movie filmed with the aesthetics of a 1980s Visage video. There’s loads of 80s fashion: heavy dry ice fog treatment in interior spaces, hair gel, crimson lipstick, caked-on eyeliner, a scene with pigeons fluttering (in slow motion), and – of course – a prominent synth soundtrack. The aesthetics reminded me of Blade Runner (1982) meets Cat People (1982).
The Hunger is considered by many to be an iconic 80s post-punk classic, and a Goth favourite, complete with an opening club scene with the band Bauhaus playing ‘Bela Lugosi’s Dead’. The plot essentially takes the goddess from She into a vampire story. Her hedonistic lust for blood (to remain young) has connotations with drug addiction. The 80s references are fun and the leading actors do a great job (David Bowie, Catherine Deneuve, and Susan Sarandon).
Thematically, there’s the idea of a violent and pampered lifestyle without moral impingement, the promise of everlasting youth versus imminent mortality, idle leisure and carnal lust. Like any good 80s MTV pop video, there isn’t much going on beneath the heavily stylised surface.
Notes
- The soon-to-air Apple TV series, Silo, is based on the self-published trilogy of Amazon Kindle books by Hugh Howey. Wool, and Shift, the first two novels were assembled from separate novellas, while Dust was written as a single, continuous story. The story takes place in a post-apocalyptic world. The surface of the earth is toxic and humanity has moved underground, into the silos. I read the novels a while ago. While they pose some intriguing questions (Why is everyone living in the silos? What’s outside of them?), the novels reflect their origin as a series of novellas and can be rambling in places. They could have done with an editor to shape them. My hunch is that the film format (which the film rights were originally purchased for) might have worked better, because this kind of story hinges on a big revelation. Once that’s gone the story looses any sense of tension. I suspect that the success of the series will hang on Rebecca Ferguson’s character. I say character rather than performance, because an actor can only do so much – there’s a lot hanging on the film-script. Hopefully the series will be more Severance than Y: The Last Man, and it won’t turn into a sitcom in a bunker.
- It’s strange how you can have a fairly decent grasp of a writer’s work without ever having read any of their writing. This is even more true when they are part of the so called ‘cultural milieu’. I’ve had my own preconceptions about Graham Greene’s novels, and now that I’ve actually read four of his novels I’m struck by the uncannily accuracy my assumptions. His writing has delightful literary touches, brilliant philosophical life-observations, and a nuanced approach to character building. These elements combine to create the hallmarks of a Greene story: the personal meets the political, nuance, intelligence, and an accessible read. But, as much as I admire Greene’s skills as a writer… I’m not really a fan. The main issue for me, as a reader, is the slow pace. Part of this reflects the fact that his fiction lives in a grey area between genre type plots (‘entertainments’ as he called them) and literary fiction. His novels often feel like they want to lurch into genre fiction, but they never do. The End of the Affair, for example, (which starts with a superb opening line that’s worthy of Wittgenstein) is a protracted confession about personal inadequacy rather than a conventional romantic drama. The Honorary Consul isn’t so much a kidnap triller but a series of character portraits of people facing moral dilemmas. Greene’s eye is on the battle of beliefs and values. His accessible lit-fit style echoed his influences, from popular authors like H Rider Haggard, Robert Louis Stevenson, and John Buchan, to literary writers like Conrad, Ford, James, and Proust. Greene is interested in internal worlds, and moral ambiguity.
- Dangerous Edge: A Life of Graham Greene is a documentary about Greene’s complicated and often dysfunctional life. His problems appear to stem from his traumatised childhood, and they seep into his characters. Sometimes, in The End of the Affair, for example, his own life became the undisguised subject matter of his writing. He was a writers’ writer, suffering for the art, but also ambivalent about the craft. Ultimately, he recognised the moral inconsistencies within himself, and in the wider world.
- Set in a 1980 USA, Armageddon Time is a coming of age story that explores aspects of childhood alienation, friendship, adult role models, hostile school environments, social division and racial prejudice. It’s a predictable, but sufficiently enjoyable watch.
- The film Living is a straightforward, well executed story with attractive, albeit highly controlled cinematography. The storytelling sticks to a ‘keep it simple’ philosophy, with a slow pace and downbeat tone. If you like films that have a theatrical quality about them, then you’ll love this, but if you prefer a more naturalistic style then you might be frustrated.
- Mark Kermode’s review of Avatar: The Way of Water was excoriating. Is it that bad? I have three distinct takeaways from watching the film. Firstly, at three hours twelve minutes, it’s far too long. Secondly, while it is technically impressive – with state of the art CGI / motion capture / performance capture – it’s more of the same… but turned up to eleven. Thirdly, the film sacrifices story flow for a series of computer game-like action sequences. After two hours, the visual hyperbole becomes tiresome. In the third hour, the film switches into action mode and manages to save itself from 120 minutes of tiresome world building with exciting action. If only the first two hours had been shorter.
- This J G Ballard interview from 1977 is interesting, and there’s the nostalgia factor of the old Thames television logo at the start. In the interview he talks about being a science fiction writer and writing science fiction, which is slightly strange because I will always see him as a literary fiction author (even when he was writing ‘science fiction’).
- This episode of the The Rachman Review podcast asks: ‘Is there such a thing as a rules-based international order?’ Or, is there only raw geopolitical power?
- Paul Morland on the Off Script podcast. This episode is titled ‘Britain’s demography and the truth about immigration’. His slightly weird, The Stepford Wives-ish conclusion… start having three children families or accept the need for immigration.
- The Economist’s Babbage podcast asks, ‘How worrying is generative AI?’ This is a fascinating look into how generative AI actually works, and the employment implications for white collar workers in the future.
- Meanwhile… here’s something important. It’s a YouTube video of a weird 1970s style onion sandwich (yes, really). Originally created by TV chef James Beard, it’s been recreated in this TV clip by chef Jacques Pépin. What happened in the 1970s? Strange doesn’t even begin to describe it.