Since launching Damonza in 2012, my aim has always been to increase the quality of the book covers we offer our authors. That goal has driven every expansion of our toolkit over the years.

We’ve always used digital tools to edit and manipulate illustrations, photographs, fonts and other design elements. We don’t paint our covers or draw them with pencils, but we may use a Photoshop filter to recreate those effects. That’s simply how the art of cover design works.

When AI image generators first appeared (Midjourney, Stable Diffusion, DALL·E), I was concerned they’d make my team redundant. Just type “design me a book cover,” and hey presto, it’s done.

But cover design is so much more than producing a decent-looking image. The more I explored generative AI tools, the clearer it became. They’re not designers. They’re tools. Powerful ones, but still tools. You still need experience, taste, genre fluency, and a sense of what works in the real world.

Generative AI is just that: another tool, albeit a game-changing one. Digital artists didn’t replace traditional artists. Photoshop didn’t replace designers; it evolved them. Generative AI is simply the next evolution, and quite literally part of Photoshop itself now, with tools like Generative Fill baked in by default.

Major platforms like Shutterstock include AI-generated images in their libraries. Canva lets authors generate AI artwork directly. AI-generated elements are already on thousands of book covers. You’ve seen them. You may have bought one.

So, the genie’s out. The question is: how do we use it responsibly?

Two different things people call “AI”

It’s worth separating two things that often get lumped together.

Generative AI creates new imagery from a written description. Tools like Midjourney, Adobe Firefly, ChatGPT/DALL·E, Google’s Nano Banana (Gemini), and Freepik work this way. We can use them to generate a specific element, such as a creature, a setting, or a texture, that we then composite into a larger design.

AI assistance helps a designer do something they’re already doing, faster. Modern Photoshop is full of these tools: extending a background to fill a print bleed, removing a distracting object, upscaling an image, sharpening fine detail. None of these invent new imagery. They speed up tasks designers have always done by hand. They’re as much a part of contemporary design work as the clone stamp or the healing brush.

The simple test we use: could a designer have done this with standard Photoshop tools given enough time? If yes, it’s assistance. If it’s genuinely impossible without AI, it’s generative.

The ethics of AI-generated images

When AI image generators first launched, there were legitimate questions about training data, especially copyrighted material. Initially, I shared some of those concerns. But over time, I’ve come to see that the ethical line isn’t in the tool. It’s in how it’s used.

At Damonza, we’ve designed over 15,000 book covers. I’m sure many of them have been swept into AI training sets. I could feel victimized, but honestly? That’s not how creativity works. Artists have always learned from what came before. So has AI. The difference is speed.

And let’s not forget that most book covers aren’t fully copyrightable in the first place. If you’re using pre-existing fonts and stock images, as almost every cover does, you’re working with licensed elements. AI-generated assets are no different in that respect. They’re just new material, made on demand, under clear usage terms.

We don’t mimic other artists. We don’t prompt with anyone’s name. And we never pass off AI’s raw output as finished work. We use it as a creative ingredient, carefully shaped, styled, and integrated by professional designers, just like any other asset in the toolbox.

We can now make your cover dreams come true

Relying only on stock photography was always a bit limiting: representation gaps, weak genre specificity, and the same stock faces appearing on dozens of books. AI gives us something stock never could. Specificity, originality, and a broader range of creative control.

Now, we can bring to life exactly what an author imagines, whether that’s a very specific character, moment, or setting, without compromising quality or consistency. AI lets us build bespoke cover elements that wouldn’t otherwise exist. We’ve even been able to eliminate our “character creation” fee because we can now produce fully custom people, scenes, and environments far more efficiently.

“Where once we were limited to stock elements, we now have literally infinite options.”

Some briefs simply can’t be executed well using existing imagery. The character is too specific. The scene is too unusual. The combination of elements has never been photographed, painted, or illustrated. A stock-photo approximation would mean settling for something that’s close but not right.

Our job is to deliver the best possible cover for your book. Sometimes that means stock photography and traditional design. Sometimes it means commissioning custom art. And sometimes it’s generative AI, because it lets us create exactly the element your book needs rather than the closest available substitute.

When we choose AI for a particular element, it’s because nothing else gets us to the standard your book deserves, not because it’s cheaper or quicker. The composite, the integration, the design judgement: those still take the same time and care they always have.

How we use AI responsibly

When generative AI is used (and it isn’t every time), it’s under the following strict conditions:

• Only separate elements are generated, never the full cover, and never the typography.
• AI elements are combined with royalty-free stock images and/or custom design work.
• No real people or artists are referenced in prompts.
• All AI-generated assets are created privately, not publicly searchable or reusable.
• All AI images are generated with commercial licenses appropriate for book use.
• Every cover is reviewed and corrected by a human designer to fix the common AI errors: hands, faces, symmetry, patterns, anatomy, and the other tells that give raw AI output away.

Your choice on AI use

Our order form includes an opt-out checkbox, so authors can choose to avoid generative AI entirely. If you opt out, that decision is final on our side: we will not use generative AI on your cover, on any print extensions (paperback wraps, hardcover wraps), or on any marketing collateral we produce for your title.

The only exception is the rare situation where, after looking at your brief properly, we genuinely don’t believe we can produce a cover that meets the standard you deserve without it. In that case we’ll tell you before any design work starts, and you can decide what to do: stay with the no-AI approach (we’ll do our best within those limits), allow AI for specific elements, or cancel the project for a full refund. We will not start work and then surprise you with this conversation later.

And if you allow generative AI on your cover, we’ll let you know in our delivery message whether we used it and where, so you always know exactly what went into your design.

A note on Amazon KDP

If your cover uses generative AI elements, Amazon KDP requires you to declare this when you publish. The declaration applies to generative AI only. AI assistance (Photoshop’s standard tools, upscaling, background extension, and so on) does not need to be disclosed.

This declaration is private between you and Amazon. There’s no public label on your book listing, no badge in the Kindle store, no tag in search results, no flag on the product page. Readers see your book exactly the way they’d see any other book. The disclosure exists because Amazon wants internal data on how AI is being used across the platform. It’s not a consumer-facing warning.

We design covers for many bestselling authors whose covers contain generative AI elements. Their books sell exactly as well as anyone else’s, because what readers respond to is whether the cover works for the book, not how individual elements were sourced.

Want to know more?

We’ve covered every angle of this discussion in our full article series:

The Legal and Ethical Use of AI in Book Cover Design
AI in Design: Just Another Tool in a Cover Designer’s Toolkit
Addressing Concerns: Understanding and Overcoming Hesitation
Boosting Creativity: How AI Enhances Book Cover Design

I also had the chance to discuss our approach to AI in cover design with Joanna Penn on The Creative Penn Podcast:
Using AI Images In Your Book Cover Design Process With Damon Freeman

For grounded perspectives from elsewhere in the indie author world, I recommend this follow-up conversation between Joanna and designer James Helps from Go On Write:
A Creative Approach to Generative AI in Book Cover Design with James Helps

And this earlier podcast featuring Joanna and others in the indie space:
Generative AI and the Indie Author Community

And as always, if you have questions, I’m just an email away: damonza@damonza.com

Thanks,

Damon