Book Box: The death of the author?

DSpike Reader, the likes of many writers, I feature suit possessed(p) with AI. Give me a piece of recent writing and my brain begins to whir, trying to spot the ghost in the wires.I scan text for AI ‘tells’: the notorious em dash, the rule of three (”Not A. Not B. Just C.”), and the tell-tale words that AI loves like “delve,” “tapestry,” and “quiet”—the quiet grief of adult friendships, the quiet erasing, the quiet…..There go the individual voices with their quirks, the little roadside dhabas serving aloo parathas, butter chicken and noodles. Instead, everywhere you have the smart, sleek franchises of large language models, with their uber-readability and their short, snappy sentences.Storytellers have predicted this. Go back in time, for instance, to 1953, to sci-fi writer Isaac Asimov’s short story, “The Monkey’s Finger.” Here, an LLM-like monkey who has been fed the world’s greatest texts demonstrates how it can do better than the writer, who is then pressured by his publisher to change the ending of his story.In The Great Automatic Grammatizator by Roald Dahl, engineer Adolph Knipe invents an LLM-like Grammatizator which is fed “plots,” “tropes,” and “styles of writing,” and then programmed to come up with different stories. By the conclusion, “half of all the novels and stories published in the English language were produced by Adolph Knipe upon the great automatic grammatizator.”Last year, Nnedi Okorafor, the Hugo and Nebula Award-winning Nigerian American writer, published Death of the Author. It tells the story of Zelu, a disabled Nigerian American writer in her early thirties. In another thread, we have the story of a universe of robots. I loved the book for the way it looks at the nature of storytelling as it moves between these two threads, one human and one machine, looking for a solution and finding one in a surprise twist.Where Asimov and Dahl feared the machine would replace us, Okorafor imagines possible solutions. And now the literary world is in a furore after writer after writer admits to using AI in their creative process. Earlier this year, romance writer Carol Hart featured in a story that went viral, as she proudly stated she uses AI to write more than 200 romance novels a year, many under pseudonyms.A few days ago Olga Tokarczuk caused a furore by admitting to using AI in her writing.“Often I just ask the machine, ‘darling, how could we develop this beautifully?’” the 64 year old Nobel prize-winning Polish writer reportedly told an interviewer.How do we as readers respond to this? There are tools to detect the use of AI, but they are tricky, because AI mimics human patterns so well that detection is almost impossible. AI is only using the same rules of human writing that humans have used for centuries—if you go looking for em dashes, for instance, you will find them in the best-known writers, including poets like Emily Dickinson.At our book club, we are meant to discuss Wifedom, an investigation by Australian journalist Anna Funder into how George Orwell used his wife Eileen’s ideas and her writing as well in works like Animal Farm without acknowledging her at all. Instead, we are discussing AI in writing, how to detect it, the ethics of it, and whether we should read authors that admit to using AI.My favourite answer to this read-or-not-to-read question comes from my friend and book clubber, Kavitha.“Perhaps it’s time to go back to reading only the previous centuries’ works, which might have only been edited, embellished, beautified, or written by unacknowledged wives,” she says.(Sonya Dutta Choudhury is a Mumbai-based journalist and Founder, Sonya’s Book Box, a bespoke book service. For all questions about life and literature email sonyasbookbox@gmail.com.)