Let’s be real for a minute: Artificial Intelligence has turned the writing world upside down. Everyone’s talking about it, especially in light of The Atlantic’s investigation into the Library Genesis data set finding that more than 7.5 million books were copied, at least in part, if not in full, by Meta and other AI companies for their AI systems. If you’re just finding out about this, the article includes a search tool so you can check to see if your book is on the illegal pirating site.
And let me tell you, as someone who makes her living writing books for other people, this has naturally been on my mind a lot.
While pirating writers’ work is absolutely not okay (my latest book doesn’t show up in the search—probably because it was published too recently—but a book I edited with my advisor in grad school is there), burying our heads in the sand or demonizing AI isn’t the answer. We aren’t going to stop this train.
As with all big technological changes, we have two options: panic or adapt. I’m choosing the latter, and I suggest you do the same (whether you’re a writer or not).
- Next week (April 24), I’m holding a webinar with Ravit Dotan, an AI adoption advisor, to discuss the work we’ve been doing together to reinvent ghostwriting. Register here!
- If you’re thinking about using AI to write a business book, you’re welcome to have a 1-on-1 conversation with me. You can schedule it right here.
- You are also welcome to a 1-on-1 conversation with Ravit about reinventing your specific profession. Sign up here!
But this does raise an interesting question: What is a ghostwriter’s role in the age of AI?
My Bionic Ghostwriting Experiment
Over the past few months, I’ve been experimenting with incorporating AI into my ghostwriting process with two book clients. My biggest concern about using AI was that I would feel like I was outsourcing my creativity.
As a ghostwriter, my most valuable skill is listening to my clients talk about what they do and figuring out their unique angle on their topic. Although the ideas I write about are not mine, I get to steer the mind meld that happens when a client feeds me great ideas. And I love that collaborative, iterative process.
Let me share what I’ve learned so far (spoiler alert: it’s been a game-changer, but not in the way you might think).
Starting with Structure: The AI-Assisted Outline
In the past, I’ve required my ghostwriting clients to bring me an outline at the start of every project. This time, we started by using Claude to help us develop a working outline. It wasn’t a five-minute process where we asked AI to “create an outline for a business book about X” and called it a day. Not even close.
Instead, getting to a usable outline took a couple of weeks of deep conversation, strategic thinking, and back-and-forth refinement. We fed Claude information about the client’s expertise, target audience, and unique perspective. We asked questions, evaluated the responses, challenged assumptions, and revised repeatedly until we landed on an angle, a narrative arc, and outline that felt right.
The key takeaway: AI didn’t replace our strategic thinking—it enhanced it. It helped us identify gaps in our logic, suggested alternative structures, and occasionally sparked ideas we hadn’t considered. But the outline still required human judgment at every step.
Where AI Fails: Full Chapter Generation
Here’s what hasn’t worked well: asking Claude (or any AI) to write complete chapters from scratch.
I’ve tried this approach, and the results have been… meh.
AI-generated chapters often sound generic, lack the emotional depth that resonates with readers, and miss the nuanced expertise that makes a business book valuable. They’re a solid C+ at best—and my clients deserve A+ work.
The Real Breakthrough: Voice Analysis and Refinement
Where AI has truly transformed my process is in voice analysis. This is where things get really interesting.
As a ghostwriter, I’ve always had to develop an intuitive feel for my clients’ voices. Before AI, this involved reading everything they’d written, listening to their podcast interviews or talks, and developing a mental model of how they communicate. It worked, but it was time-consuming, and with some clients, I arrived at this mental model too late in my writing process. So I’d have to do a lot of re-writing late in the game.
Now, I use AI to analyze my client’s writing samples and identify patterns in their communication style. Because I understand voice analysis deeply, I can collaborate with the AI to extract meaningful patterns about how my clients communicate. AI provides detailed analysis that would take me hours to compile manually.
But here’s the crucial piece: the analysis itself isn’t the end product. It’s a tool that informs my writing.
Let me share an example to illustrate what I mean. Recently, I wrote an email newsletter about a guest lecture I gave at Harvard. I started by giving Claude some basic information and asking it to draft the email. I then refined it using a process I’ve developed along with the tone analysis tool I’m calling the “Emily-ifier.”
Here’s what I got:
Claude’s Version (Formal/Professional):
“Dear Subscribers,
I wanted to share a bit about my experience giving a guest lecture at Harvard University this week. Dr. Cynthia Meyersburg invited me to speak to her Human Motivation class about entrepreneurship, authorship, and how understanding human motivation has informed my work as a business book ghostwriter.
Emily-ified Version 1 (Semi-Formal/More Conversational):
Hello everyone!
While I happily left academia behind, there are certain aspects I genuinely miss—particularly being in the classroom. Every time I have the opportunity to engage with students who’ve read the same texts I have, I’m reminded of the unique joy of those shared intellectual experiences. It’s been almost a year since I last had this opportunity (when Jill Delston invited me to speak to students considering grad school at UMSL last spring), and each time I find myself thinking, “I want to do more of this.”
This week, I had the privilege of speaking to Dr. Cynthia Meyersburg’s Human Motivation class at Harvard University – a room of about 50 bright, engaged students who had read excerpts from my book “Unwritten.” The experience gave me a chance to think more deeply about what I’ve learned about human motivation through working with my ghostwriting clients.
Emily-ified Version 2 (Conversational/Authentically Me):
Hey there!
I left academia. (Best decision ever, honestly.) But there are certain aspects I genuinely miss—particularly being in a room full of people who’ve all read the same thing I have. There’s something magical about that shared intellectual experience that just doesn’t happen anywhere else in my life.
Know what I mean?
It’s been almost a year since I last had this classroom fix (when my friend Jill invited me to speak to some students considering grad school at UMSL). Every time I get to do it, I walk away thinking, “Sh*t, I need to do more of this!” There’s something so energizing about engaging with curious minds who ask thoughtful questions.
This week, I hit the jackpot. Dr. Cynthia Meyersburg invited me to speak to her Human Motivation class at Harvard—a room of about 50 students who had actually read excerpts from my book “Unwritten.” (Surreal!) The experience pushed me to think more deeply about what I’ve learned about human motivation through my ghostwriting work.
I’m pretty sure I walked away with my next book idea. Funny how that happens when you’re not even looking for it, right?
It’s easy to see that this final version aligns most authentically with my natural voice. It uses contractions, rhetorical questions, parenthetical asides, varied sentence lengths, and even mild swearing—all hallmarks of how I actually communicate.
This process helps me make more conscious choices about voice, not just for my own writing but especially for my ghostwriting clients. It has transformed how I approach capturing their authentic communication patterns.
Why This Matters: Beyond Time-Saving
Yes, this approach saves me time, but that’s actually the smallest benefit. Here’s what’s far more valuable:
- Focus on higher-level thinking: By reducing the cognitive load of constantly monitoring voice elements, I can dedicate more mental energy to the strategic aspects of the book—structure, flow, and impact.
- Confidence in capturing voice: Both my clients and I feel more confident that we’re accurately representing their authentic communication style.
- Better collaboration: The voice analysis creates a shared language with clients about their communication style, making feedback more specific and useful.
- Consistency across book projects: This method helps me track concepts and named frameworks we invent, ensuring we use them consistently throughout the manuscript. AI acts like a memory assistant, helping me maintain coherence from the first page to the last.
- Editorial control remains critical: I’ve noticed a common mistake among non-experts trying to use AI for book writing—they accept AI outputs too uncritically. Being a skilled editor is more important than ever. I never simply accept the authoritative voice of AI without applying rigorous editorial judgment. The ability to evaluate, reshape, and refine AI-generated insights is a crucial skill that separates professional ghostwriters from those who simply prompt AI to write their books.
Finding the Balance: Human Creativity + AI Enhancement
One of my biggest initial concerns about using AI was that it would diminish my creativity or make my writing generic. I worried I might be tempted to outsource not just the writing (putting myself out of a job), but my thinking (making me dumber).
What I’ve discovered is quite the opposite. By using AI as an analytical tool rather than a replacement writer, I’m engaging my creativity in new ways. I’m thinking more critically about voice, style, and communication patterns. I’m making more conscious choices about how to structure information and present ideas.
I still control the writing process entirely. The words are mine (or more accurately, a blend of mine and my client’s). The ideas come from deep conversations with experts who have spent decades honing their craft. AI simply helps me identify and implement patterns that make the writing feel authentic.
The Future of Ghostwriting
In the age of AI, a ghostwriter is still fundamentally what we’ve always been: a bridge between expertise and audience, a translator of complex ideas, and a partner in bringing important work into the world. We just have some powerful new tools at our disposal.
What will (and must) change is how we work. The ghostwriters who thrive will be those who embrace AI as an enhancement to their process rather than a threat to it. We’ll spend less time on mechanical aspects of writing and more time on the strategic and creative elements that truly matter.
I’m excited to see how this continues to evolve. If you’re working on a book or considering one, I’d love to hear how you’re using (or not using) AI. And if you’re curious about how this bionic approach might help bring your big ideas to life, AI ethics expert, Ravit Dotan and I are hosting a webinar at noon on April 24th to talk more. You can register for free right here!
Image by @kjpargeter at freepik