From Edie: Confused about AI? Learn the differences between generative AI, editing tools, and research tools—and what every writer needs to know.
Understanding AI for Writers: Generative AI, Editing Tools, and Their Differences
by Edie Melson @EdieMelson
Artificial intelligence is everywhere these days. And if you're a writer, you've probably heard more opinions about it than you ever wanted.
Some people think AI is the greatest thing to happen to writing since the invention of the computer. Others believe it's going to destroy the publishing industry.
Because here's something a lot of we writers don't realize.
Not all AI is the same.
And that distinction matters, especially when we're preparing manuscripts and book proposals for publication.
I've been teaching writers how to put together professional book proposals for years. Recently, I've added something new to the process: an Author AI Disclosure Statement.
Before we talk about writing that statement, though, we need to understand what we're actually disclosing.
AI Isn't Just One Thing
When someone says they've used AI, what does that mean?
Did they ask ChatGPT to write an entire chapter? Did they use Grammarly to catch a misplaced comma? Did they use Lightroom to reduce noise in a photograph?
Those are very different activities. Yet they're often lumped together under the same heading.
Let's look at four types of AI writers are likely to encounter.
1. Generative AI
This is the type of artificial intelligence getting most of the attention. Generative AI creates content in response to instructions, commonly called prompts. That content might be text, images, audio, or video. Tools such as ChatGPT, Claude, Gemini, and Adobe Firefly offer generative AI capabilities. For writers, this might mean asking AI to brainstorm possible plot twists, suggest book titles, develop a marketing strategy, or write an entire scene.
Notice something important about those examples. Brainstorming possible plot twists isn't the same as having AI write the scene. Both involve generative AI, but the way the technology contributes to the finished manuscript is different. And that's a distinction publishers may care about.
2. Predictive AI
Predictive AI analyzes patterns and uses them to anticipate what might happen next. You've probably used it without giving it much thought. Think about the suggested words that appear when you're typing a text message. Or the way your email program identifies messages that might be spam.
These are familiar examples of predictive technology.
Most writers aren't going to need to explain every instance of ordinary predictive text in a book proposal. But understanding the distinction helps us avoid treating every AI-powered feature as though it generated our manuscript.
3. AI-Assisted Editing
This is where things can get confusing. Programs such as Grammarly and Microsoft Editor can identify grammatical problems, suggest clearer wording, and help us evaluate existing text. But some editing programs also include generative AI features that can rewrite entire passages.
Those aren't necessarily the same thing.
For example, there's a difference between a program identifying a run-on sentence and having that program rewrite three paragraphs of your manuscript. The same distinction applies to photography. I use Adobe Lightroom in my photography workflow. AI-powered noise reduction is very different from using generative AI to create or replace portions of an image. Both involve AI, but they accomplish different tasks.
And this brings us to something I want every writer to understand.
The name of the program isn't nearly as important as what you asked it to do.
4. AI-Powered Research and Analysis
AI can also help us organize information, identify possible research directions, and explore topics we might otherwise overlook.
For example, a nonfiction writer might use AI to generate questions that need to be answered during research.
A novelist might ask about occupations, historical settings, or technical subjects relevant to a story.
An author preparing a book proposal might use AI to brainstorm possible comparable titles.
These can be useful starting points.
But—and this is a BIG but—AI can be wrong. It can invent sources, misrepresent information, and confidently provide answers that aren't true. I've seen enough inaccurate AI-generated information to know that we cannot afford to skip verification. If you use AI to suggest comparable books, make certain those books actually exist. Check the publication dates, authors, publishers, and other details. If you use AI for historical research, verify the information through reliable sources. And for those of us writing for the Christian market, we need to be especially careful with Scripture references, theological interpretations, and quotations.
AI is not a substitute for studying God's Word or checking our sources.
What About AI-Generated Images?
This is another area writers need to consider. Perhaps you've used AI to create a possible book cover. Or you've generated illustrations, maps, character images, or promotional graphics. Maybe you've used AI to make substantial changes to an existing photograph. Those uses may also need to be disclosed, depending on the project and the publisher's requirements.
And there's an additional issue: rights.
We can't automatically assume that because we generated an image, we own exclusive rights to it or that it qualifies for copyright protection. The rules surrounding AI-generated artwork can be complicated, and individual publishers may have their own restrictions. A cover mockup created for brainstorming isn't necessarily the same as finished artwork intended for publication. But it's important to know the difference and be transparent about what you've created.
So What Does All This Mean for Writers?
It means we need to pay attention to how we're using AI. I'm not suggesting we should panic every time a writing program offers a suggestion. I'm also not suggesting that every use of AI is appropriate for every publishing project. What I am saying is that we need to understand our tools well enough to explain how we've used them.
That's part of being a professional writer.
An AI disclosure statement isn't necessarily required by every publisher or agency. Policies vary, and some publishers restrict uses of AI that others may allow.
But being prepared to explain your process is a wise step. And if you're working on a book proposal, now is a good time to think through these questions.
- Did you use AI to brainstorm?
- Did you use it to edit?
- Did it help you research or develop marketing ideas?
- Did it actually generate words or images that became part of your finished project?
Those answers will help you create an accurate disclosure statement.
Tomorrow, we'll take what we've learned and put it into practice. I'll show you how to write an Author AI Disclosure Statement that can accompany your manuscript and book proposal.
Because the goal isn't to prove we never use technology. The goal is to be honest about our process and professional in the way we present our work.
Now it's your turn. What questions do you have about AI? Be sure to leave your thoughts—and your suggestions in the comments below.
Don't forget to join the conversation!
Blessings,
Edie
TWEETABLE
She and husband Kirk have been married 44+ years with three grown sons and four grandchildren. They live in the foothills of the Blue Ridge Mountains and can often be found with their dog hiking the mountains. Connect with her at www.EdieMelson.com and through social media.
No comments:
Post a Comment