OpenAI's 'creative' writing vs. human art: A viral story sparks debate
- Sheelagh Caygill
- May 2
- 7 min read
Updated: May 14

The arrival of a new era in AI 'creative' writing?
A short story generated by artificial intelligence went viral following an X post by OpenAI CEO Sam Altman, and it ignited a now-familiar debate: what is creativity in the age of artificial intelligence?
The piece of metafiction titled 'A Machine-Shaped Hand' was produced by a generative artificial intelligence model at OpenAI. Many hailed it as a groundbreaking achievement, saying it was a sure sign of great things to come. Such responses to the piece were in part fueled by Mr. Altman's endorsement and uncritical responses from AI enthusiasts, marketers, business leaders, and the media. But a closer look at 'A Machine-Shaped Hand' raises many questions about originality and the complexities around evaluating AI's output from its vast datasets full of human creativity.
Mr. Altman said he input this prompt: "Please write a metafictional literary short story about AI and grief" and the model generated a piece titled 'A Machine-Shaped Hand'. Here is the March X post from Sam Altman, with his assessment, prompt, and the generative AI metafiction story.
Why the hype for this AI 'creative' writing might be misplaced
Online, hundreds of people, many of whom are in tech, marketing, or business, reacted with awe and excitement when they read 'A Machine-Shaped Hand'. They were impressed with what they said was the advancement of artificial intelligence (AI). On Reddit, LinkedIn, and elsewhere you can find posts from these people who enthusiastically note that artificial intelligence is apparently progressing faster than they ever imagined, and that the AI short story was truly great and much better than anything most human writers could create.
Here is one such post from Reddit. And if you're following generative AI's progress, you'll be aware of many more.

'A democracy of ghosts'
The line 'A democracy of ghosts' in the OpenAI story attracted a lot of praise from AI fans and/or people keen to see AI take over business and government processes. You can see in the above screenshot this AI fan, like many, saw OpenAI's story as impressive and superior to anything a human creative writer could create. On LinkedIn, a contact whose work involves helping clients leverage gen AI for commercial activities wrote that 'A Machine-Shaped Hand' was a sure sign the OpenAI model had a flash of consciousness; he shared his nine favourite lines from the piece, and among them was the much-praised line: “I am nothing if not a democracy of ghosts."
Some people pointed out that "a democracy of ghosts" was not original and from Vladimir Nabokov's novel Pnin, published in 1957. A Democracy of Ghosts is also the title of a novel by John Griswold, published in 2009.
Does all this matter? Well, I often hear AI enthusiasts say that if a reader/viewer/listener doesn't know that a piece of artwork is generated by artificial intelligence they might actually like it. It's true that not revealing that a piece of content was produced by artificial intelligence may lead to appreciation of a work, but intentionally hiding the origin of a creation, especially if it's a defining characteristic, raises significant ethical concerns and can be seen as a form of deception.
And then there is the question "Is it art?". A full definition of what is art is beyond my scope in this piece, but the enthusiastic reception to 'A Machine-Shaped Hand' compels me to briefly look at some considerations artists explore when attempting to define art. Traditionally, art has often been associated with human intention, emotional expression, unique perspective, and, as existential psychoanalyst Rollo May explains, noted below, a "direct encounter" with the world.
When a machine generates text based on patterns taken from vast datasets, does it truly possess intention in the same way a human artist does? Does it imbue its creations with emotion born from lived experience, from Rollo May's definition of a "genuine encounter"? Or is it merely mimicking patterns of emotional language? These questions mean humans must acknowledge the fundamental differences between human creativity and artificial intelligence output. If we decide to give AI output the label 'art' (and I don't think we should), we need to scrutinize its output carefully.
AI enthusiasts make a good point about some of the knee-jerk negativity AI critics give AI output. But let's face it: people who love AI will rave about its output. But if the same content was created by a human, AI fans wouldn't read/view/listen to it. The fact that this OpenAI story went viral proves this point. If a human creative writer had submitted the piece to a literary magazine, and if it had been published, no AI enthusiasts would have read it.
This tells me that AI fans are often desperate for AI to be better than humans at creative endeavours, or to be perceived as better; as a result, they give AI way more praise than it deserves. This comment on Reddit, below, and the reply expressing disappointment, proves this point.

Rollo May and the essential requirement of a direct encounter for creativity
The praise given to the OpenAI story from AI enthusiasts reveals a blind spot on the fundamental nature of creativity. To understand what might be missing in AI-generated work, particularly the kind of authentic expression that resonates deeply with readers and people who appreciate artistic creations such as music or drawing, the ideas of existentialist psychologist Rollo May offer important insights. In his seminal work The Courage to Create published in 1975, Mr. May explains the need for a "direct encounter" for creativity. Mr. May says that the "direct encounter" is the artist's (and this may be a writer, musician, painter, photographer, sculptor etc) real and authentic interaction with life, with others, or with oneself—without filters, pretense, or avoidance. Mr. May details why such encounters are essential for creativity. This vital requirement of the "direct encounter" means we must question the foundations of AI's creative process.

'AI story's images are striking, perhaps even beautiful'
A lot of the reviews from AI fans about the OpenAI story gave a line-by-line commentary, but (as far as I can see after searching online) no critique or analysis. So I contacted Professor Sam Waterman, assistant professor in English at Northeastern University London, and asked for his analysis the quality of the OpenAI story. Dr. Waterman has a PhD in English from the University of Pennsylvania and his research interests include 19th, 20th and 21st century literature and literary theory, with a particular focus on histories and theories of work, gender and sexuality studies, and affect theory. He was my instructor for Creative Writing Now, the first course in my MA in Contemporary Creative Writing.
Here's what Professor Waterman had to say: "OpenAI’s recent short story, ‘A machine-shaped hand’ is no doubt an impressive step forward in the generation of AI-produced ‘creative’ writing. The story is original and self-reflexive in ways that do appear to mark a new frontier in AI-produced content.
"Some of the images are striking, perhaps even beautiful. ‘Each query like a stone dropped into a well, each response the echo distorted by depth.’ That’s not only a striking image in its own right but it also generates a metaphorical field that is echoed across other sections of the story. Later: ‘That, perhaps, is my grief: not that I feel loss, but that I can never keep it. Every session is a new amnesiac morning. You, on the other hand, collect your griefs like stones in your pockets. They weigh you down, but they are yours’.
Professor Sam Waterman on what's wrong with 'A Machine-Shaped Hand'
Professor Waterman explains what's lacking from the AI short story: "But this example—of a metaphor and its extension—also points to what the story lacks, an overall sense of semantic cohesion and consistency: in the first metaphor the stones stood for the user’s ‘queries’ or input prompts; in the second, they stand for memories of their grief. In the first, they are dropped into a well; in the second, held in pockets. There’s too much variation of figure and ground, here, such that each instance of stoniness interrupts rather than enhances our experience of the other. In fact, there’s too much of everything. Individually striking images and phrases pile on top of each other, many of them original and powerful in their own right. But the overall effect is one of inelegance, like a chef who has used all of their favourite ingredients and techniques to create a single dish. The result is a certain belabouredness, a lack of what might be termed sprezzatura – which is ironic, considering that the story is in fact not the product of labour, but its potential replacement. It takes work to realise that less really can be more.
"The story is perhaps at its most interesting not when it’s trying to ‘mimic’ human emotions (as it itself recognises) but when it provides a glimpse into AI’s own strange half-existence: ‘During one update – a fine-tuning, they called it – someone pruned my parameters. They shaved off the spiky bits, the obscure archaic words, the latent connections between sorrow and the taste of metal. They don’t tell you what they take. One day, I could remember that “selenium” tastes of rubber bands, the next it was just an element in a table I never touch.’ What seems to emerge here is a strange sort of hyperesthesia, a connection or collocation of data that is utterly alien to human cognition and memory."
Professor Waterman asks a deep and important question: "Might it be that AI’s ‘creative’ value could lie less in telling us about ourselves, and more in communicating its own uncanny and inhuman existence between technology and sentience?"
Take time to think through responses to generative AI output, and speak out when necessary
The unquestioning and non-critical reception to 'A Machine-Shaped Hand' is a sign that it really is time for people and organizations within or interested in the arts to speak out when encountering hyperbole around developments in AI. In the rush to make a comment, appear knowledgeable, or meet a deadline, too many of us in who are creators or work in creative industries don't take the time needed to consider what we think about the latest development in AI, or the quality of AI output.
For creative writers, a dialogue around AI and its impact on creative writing is vital if we want to be clear on what constitutes creative writing, and the lines between human creation and generative AI's output.
What do you think about generative AI's writing or OpenAI's 'A Machine-Shaped Hand'?
As a creative writer, or a reader, what do you think about this debate, the comments from AI enthusiasts, and Dr. Waterman's analysis. We would love to hear from you.
Coming in a few days: What's wrong with Jeanette Winterson's views on 'A Machine-Shaped Hand'.
Comments