AI in literary criticism, without ceding your judgment
Structure, polish and publish your reviews faster, yes, but never by letting an AI sign your opinion in your place.
Press reviews, reader reports for an editorial committee, advance copies received before publication: you work on texts that are not yours and, often, not yet public. Pasted into consumer AI without a data processing agreement, they leave your control before the book is even out.
The point is not to ban AI from your desk, it is already there, but to fix what it prepares and what remains yours: the reading, the judgment, the signature.
- Hours lost structuring reviews, reader reports and reading sheets, taken from reading time
- Embargoed manuscripts or unpublished proofs pasted into public AI, beyond any control
- Quotes too long or poorly credited, with copyright engaged under your signature
- A review copy or a conflict of interest left undisclosed, and the review’s credibility with it
See for yourself
A prompt structured in 5 blocks: Instruction · Role and audience · Data · Constraints · Format. An unframed AI invents; the frame stops it.
In a workshop, we adapt these prompts to your practice, each with its safeguards: no embargoed text in a public AI, no opinion written in your place.
Critique courte presse / blog (300 à 500 mots)
The short format under control, hook, arguments, closing line, without sacrificing your angle.
See it in Chaptal →Avis négatif argumenté et respectueux
An unfavorable review grounded in the text, never in the person, that you can sign without blushing.
See it in Chaptal →Réponse à un auteur ou à un éditeur qui conteste une critique
A composed reply that defends your review on the facts without inflaming the exchange.
See it in Chaptal →30 use cases cover your line of work in the Chaptal catalog, out of 750 in the corpus · each one states what it produces, what it expects from you, and what it refuses to do.
« Le Collaborateur, augmenté »
You learn to frame AI on your real texts, reviews, reader reports, replies to publishers, without handing it the judgment.
See the training →AI usage audit
A survey of what already circulates in your newsroom or publishing house, from review copy to publication, and the framework to hold it.
See audit & advisory →Foire aux questions
Can AI write my review for me?
It can structure an outline, rephrase, check the coherence of an argument. Judgment requires having read the work and standing behind it under your name; the catalog prompts place AI before and after your opinion, never in its place.
Can I paste a manuscript received as a review copy into an AI?
Not into a public AI without a data processing agreement: the text is not yours and is often not yet published. You work from your reading notes rather than the full text.
Will my readers see that AI helped me?
If AI writes the opinion, the text turns smooth and interchangeable. If it prepares the structure while the judgment, the examples and the voice remain yours, what people read remains your review.
What about quotes and copyright?
Swiss law allows short quotation for the purpose of criticism, with the source credited; unframed AI lengthens excerpts without a second thought. The dedicated card in the catalog sets these rules before writing.
Do I have to disclose a review copy or paid content?
Yes, that is what protects your credibility: a book received, a personal tie or paid content is flagged to the reader. A card in the catalog lists what must be disclosed, and how to word it soberly.