Can AI keep track of a series bible across books?
- Models do not remember previous conversations unless a tool stores them.
- A series bible must live in a document, not in a chat history.
- Structured facts beat prose: names, dates, rules, one line each.
- Update it as you draft, or it silently goes stale and misleads.
- AI is good at extracting a first draft of the bible from finished books.
Not by itself. A model has no memory between sessions, so anything it knew about book one is gone by book three unless a tool is storing it and supplying it. What AI does well is help build and maintain the bible: extracting names, dates and established facts from manuscripts you have already written.
Chapter i·Why it matters
Series continuity errors are the most expensive kind, because readers who have invested in three books notice and mind. Writers assume the tool that helped with book one will remember it, which is exactly the assumption that lets a character eye colour change between volumes.
Chapter ii·What to include
- A single canonical bible document, versioned, outside any chat.
- Characters with the details that recur: age, appearance, speech habits.
- World rules and their limits, stated as things that cannot happen.
- A timeline across books, not just within one.
- An update habit: revise the bible when the draft contradicts it.
Chapter iii·Example
A trilogy writer has AI extract every stated fact about her cast from books one and two, checks the list, and keeps it as the bible she pastes in when drafting book three. The extraction took an afternoon and caught two contradictions she had already published.
In WriteLoom
Keep the plan alive while you draft
WriteLoom keeps characters, world entries and timeline in the project itself, and series management carries them between books rather than between conversations.
Not ready for a workspace yet? Get the free six-part series on winning the writing game: finishing drafts, editing in the right order, querying with real math, and launching on a runway.