What happens when an author admits to using AI in a prize-winning novel? It just happened in Japan and people are getting salty and surprised. Read on to find out more.
Would you or your organization like to take our workshop on bootstrapping your AI writing career? We offer comprehensive training for PR professionals, marketers, and journalists. Reach out and let’s help turn AI into your ally.
The fog rolled in thick and heavy, shrouding the dimly lit streets of Barksville in an eerie silence. In the heart of this canine-infested town, a chilling crime had taken place. A murder most foul, and the only one with the wit and the will to solve it was a rather unconventional detective — a six-foot-tall talking dog named McGruff.
With a trench coat that seemed almost comically oversized, McGruff strutted down the cobblestone streets, his oversized paws making a distinct rhythm with each step. He was no ordinary hound; he had a nose for justice and a bark that sent shivers down the spines of even the most hardened criminals.
As the street lamps flickered and the moon played hide-and-seek behind the clouds, McGruff knew he had a mystery to unravel. The victim was a prominent member of the Barksville community, and the list of suspects was long. McGruff’s deep, gravelly voice echoed in the night as he muttered to himself, “Time to sniff out the truth, one clue at a time.” And so, the tall, talking dog began his investigation, determined to unravel the secrets hidden in the shadows of Barksville and bring the perpetrator to justice.
How do you like the beginning of my new novel, The Postman Always Rings Twice But I’ll Bark At Him For Hours? Using a single prompt, I was able to generate three paragraphs of passable writing, suitable for a children’s book about a anthropomorphic dog detective.
But is that writing mine? Is it ethical? And should anyone, novelists especially, use chunks of AI writing verbatim in their writing?
That question came to the fore when Japanese author Rie Kudan admitted she used ChatGPT to create part of the conversation between an AI and a young woman in her book set in the near future, “The Tokyo Tower of Sympathy” (“Tokyo-to Dojo-to”).
The book itself won the Akutagawa Prize (the judges deemed it “flawless.”) But there was, in a way, a flaw.
Experts say AI is generative. This means that it creates novel things. I wouldn’t say that it creates novel ideas — that’s a stretch — but it can create something new that has never been seen. But it does this work based writing that it collected all over the Internet. That McGruff story above? It’s novel, but it was pieced together like Frankenstein’s Monster from multiple sources.
Did Kudan plagiarize? Absolutely not. She did the equivalent of the cut up technique, creating a new work out of work that already exists. Bob Dylan admits to using this technique as does Jeff Tweedy of Wilco. That the cut up slices of text self-assemble into grammatically-correct writing is a bonus.
But when unscrupulous writers begin using generative AI writing as the basis for their books, who wins? The grey slime of AI-generated content is already filling up the web, and Amazon’s book store is next.
I disagree with many that writers should disclose when they are using AI to help write paragraphs or chapters. It is no different than using Word vs. a typewriter. The tool is irrelevant. But bad writing is bad writing and being lazy when using AI is a crime against creativity. Kudan’s story will become more and more familiar as AI creeps into every corner of the writer’s life. It’s up to us to temper its power with our own humanity.
The Media Copilot is hosting a meetup on Feb. 1 in New York City! Join us for an evening of conversation over drinks with journalists, PR professionals, and executives about AI’s role in media and the news. RSVP to reserve your spot.
Leave a Reply