Writing a Book with AI

how LLMs are the perfect writing buddy

Less than a year ago, like many authors who are proud of their writing prowess, I snubbed my nose up at AI. Since I started exploring if AI could be taught to meditate, my opinion of AI as a writer’s tool has completely changed.

So much so that a book I didn’t plan to write was conceived at the end of last year and I am now in the process of writing it. Noting that it will not be written by AI or written by this human but co-created, co-written, co-edited, co-produced, co-marketing with AI. I may even use ElevenLabs to create the audio version.

As part of my ‘Stacking in Tune with the Moon publishing schedule in 2026, I will be expanding on each of the 12 stages summarised here … and letting you know more and more about the book I didn’t plan to write.

---

🌒 The 12 Stages of Writing a Book with AI

(Ideation → Scribing → Refinement → Manifestation → Release)

I. Ideation & Field-Setting (Stages 1–3)

The book does not yet exist, but its gravity does.

1. The Call (Why This Book?)
Clarifying the necessity of the book.

• What is asking to be written now?

• What tension, gap, or unanswered question is alive?

• AI as mirror, not oracle.

2. The Spine (What Holds It Together?)
Defining the structural intelligence of the book.

• Central metaphor / system (e.g. the missing Arcana of the Tarot - very Dan Brown)

• What repeats, what evolves

• Early resistance as signal, not flaw

3. The Shape (What Form Wants to Emerge?)
Choosing format before content calcifies.

• Linear vs modular vs lattice

• Chapters, keys, interleaves, case studies

• Where AI assists with option-space, not decisions

---

II. Scribing & Co-Creation (Stages 4–7)

The book learns to speak through you.

4. Opening the Channel
Establishing a repeatable writing state.

• Rituals, constraints, time containers

• Using AI to hold continuity between sessions

• Protecting the authorial voice

5. Rough Scribing
Writing without premature coherence.

• Drafting fast, editing later

• AI for scaffolding, summaries, memory

• Allowing asymmetry and uneven density

6. Dialogic Deepening (Human ↔ AI)
Using AI as a thinking partner.

• Asking clarifying, adversarial, and reflective prompts

• Stress-testing ideas without outsourcing authorship

• Where AI sharpens, not softens

7. Pattern Recognition
Seeing what you’ve actually written.

• Emergent themes, repetitions, omissions

• AI-assisted pattern spotting across drafts

• Adjusting structure after content speaks

---

III. Refinement & Integrity (Stages 8–10)

The book becomes trustworthy.

8. Editorial Alchemy
Transforming raw material into readable form.

• Layered editing passes (clarity, flow, tone)

• AI for compression, expansion, and contrast

• Knowing when not to optimise

9. Voice & Coherence Lock-In
Ensuring the book sounds like one mind.

• Harmonising chapters written months apart

• Preventing AI-induced blandness

• Final authority stays human

10. Embodiment Test
Can the book be lived, not just read?

• Exercises, reflections, invitations

• Where the reader enters the field

• AI used to test reader journeys, not predict them

---

IV. Manifestation & Release (Stages 11–12)

The book leaves you and becomes a companion to others.

11. Production Without Panic
Turning the manuscript into an artefact.

• Formatting, covers, metadata

• AI for logistics, checklists, sanity

• Keeping momentum without rushing closure

12. Launch as Invitation (Not Performance)
Releasing the book as a relationship.

• Marketing as storytelling, not persuasion

• Aligning Substack, courses, meditations

• Letting the book continue its life beyond you

---

Let me know if you have any other suggestions or stages we have missed.