Artificial Intelligence represents computational systems’ capability to execute tasks usually associated with human intelligence. Generative artificial intelligence (generative AI) is a form of artificial intelligence that learns from existing data to generate new, original content including text, video, audio, images, and code. In this blog post, I will explore how authors can ethically use generative AI to generate ideas for future novels.
An important truth to keep in mind when using generative AI to create story ideas for novels is that generative AI does not inherently understand story structure. Because it generates new text that is culled from lines of existing text, it does not really know anything. For example, generative AI often repeats words too often, uses too many adverbs in dialogue tags (ex. said loudly), and moves characters from one location to another without explaining how they got there, among other narrative errors. Therefore, it is best if you understand story structure before using this tool to craft new ideas. Related to this, because generative AI pulls from existing text, it is best to direct generative AI to create an idea based on a type of story structure rather than on an author’s name. Otherwise, you run the risk of the generated text reflecting a specific author’s voice, and you might be tempted to flesh out the story idea using that voice.
With all that said, here are some valuable tips to ethically use generative AI to help craft ideas for novels:
- Use AI only as a tool that supports the creative process, rather than using it to write for you. You should define your thinking, writing, and narrative voice by what you generate, not what is generated by AI. Generating text through generative AI is prompting, not actual writing.
- If you decide to ignore the above advice and incorporate any AI-generated text in your novel, you must disclose this text to your publisher. All publishing contracts require authors to assert that their manuscripts are original to them. AI-generated text is not original and is therefore not copyrightable. As such, you must also declare AI-generated text in your application for copyright registration, and your publisher requires that information to correctly register your work’s copyright.
- You should also disclose any AI-generated content to your readers, so they will not feel duped. You may consider listing AI as an “author” in the front material or acknowledgements.
- Consider adding the Author Guild’s Human Authored Certification mark to your book to indicate that it does not contain any AI-generated text. Your readers will appreciate knowing that the book is entirely human written.
- Review and fact-check all content generated by AI systems, as you cannot trust the accuracy of any facts provided by generative AI. Generative AI might also be biased against gender, race, or socioeconomic status, so you need to look out for this to avoid perpetuating misinformation or harmful stereotypes.
- During your contract negotiations with a publisher, consider including a clause that prohibits the publisher from using your work to train AI technologies without your express permission.
Now that you are mindful of using generative AI in an ethical manner, what is the procedure for generating an idea for a novel using generative AI? First, you should create an account with a generative AI software service (for the purpose of this blog, I created an account with ChatGPT using a personal e-mail address and password). The main ChatGPT screen is just an empty search bar like you would see on a search engine website. After creating the account, I typed “generate an idea for a science fiction action novel set in Los Angeles, California, in the far future, with the protagonists being a 25-year-old Caucasian male spy and his 27-year-old Caucasian female partner” into the search bar and clicked the submit icon.
And here is the output from ChatGPT:
Title: Neon Divide
Setting:
Los Angeles, 2149. The city is a vertical sprawl of glittering megastructures rising from the Pacific coast, split between skybound elite towers and shadowy undercity slums. Corporate states have replaced traditional government, with Los Angeles acting as the capital of the Western Corporate Alliance. The skyline bristles with drones, neon-lit air highways, and digital billboards that hijack the mind with subliminal commands. Beneath the glamour, resistance movements and black-market tech thrive in the labyrinth of tunnels under the city.
Protagonists:
- Ethan Voss (25) – A rookie but gifted spy working for the remnants of a federal intelligence agency now overshadowed by corporate security forces. Calm under pressure, he specializes in infiltration and counter-surveillance. Ethan has a knack for adapting quickly but wrestles with the moral compromises of espionage.
- Maya Rourke (27) – His more seasoned partner, sharp-tongued and pragmatic, with a background as a corporate defector. Skilled in cybernetics and hand-to-hand combat, she’s haunted by her past allegiance to the very conglomerate they’re fighting against.
Premise:
When a covert operation exposes evidence that a mega-corporation plans to launch the “Neural Dominion Project”—a brainwave network designed to remotely influence emotions and decisions of the entire city—Ethan and Maya are tasked with preventing its activation.
But the deeper they go, the more they realize the project isn’t just about control—it’s about rewriting human consciousness to erase individuality. Their mission takes them from glittering sky casinos and drone-filled corporate arcologies to the flooded ruins of Old Hollywood, now a stronghold for underground hackers.
Conflict:
Ethan struggles to maintain his humanity as the mission forces him into increasingly ruthless choices, while Maya confronts old enemies who know her secrets. Together, they must outwit augmented assassins, rogue AI security systems, and double agents, all while navigating the thin line between loyalty and betrayal.
Twist:
They discover that the Neural Dominion Project was secretly initiated by what remains of the U.S. government itself—not just corporations. Both corporate and federal powers are conspiring to merge into a single techno-regime. The only hope lies in a dangerous gambit: hijacking the project and turning it against its creators—at the cost of possibly erasing their own identities in the process.
Tone & Style:
Fast-paced action, neon-drenched noir atmosphere, espionage thriller with cyberpunk aesthetics. Explores themes of identity, surveillance, free will, and what it means to be human when the mind itself is a battlefield. As you can see, this example novel premise is detailed, but since it offers only the setting, characters, general premise, conflict, twist, and tone & style, you can avoid any narrative errors on behalf of the generative AI software. This premise also lacks a clear authorial voice, so you can add your own author voice later. However, the premise offers strong, intriguing concepts that you can build upon if you wish to use it as a template for a novel. Below this premise, Chat GPT asked me, “Do you want me to flesh this out into a chapter-by-chapter outline, or would you prefer a more cinematic synopsis, like a movie pitch style?” I answered “Neither,” so ChatGPT responded, “Would you like me to expand the idea itself (world-building, character arcs, unique sci-fi tech, themes), or keep it short and punchy like a single high-concept pitch you could build from?” Either idea is ethically valid (although I would recommend reviewing the character arcs very carefully for any logical errors), so I will leave it up to you to decide. What would you do? Let us know in the comments and follow our blog for more useful advice.
By Matt Wade
Matt is an Editorial Assistant at Technica Editorial