1. Purpose and scope

1.1       This policy sets out how Dapple Grey Press Ltd ("the Press", "we", "us") approaches generative artificial intelligence ("AI") in connection with the works we acquire, publish and promote, and how we respond to public statements about the origin of those works.

1.2       It applies to all authors, illustrators, editors, designers and other contributors who work with the Press ("Contributors"), and describes our position to readers, retailers, schools and the public.

1.3       It is a statement of policy. It does not form part of any contract between the Press and a Contributor unless expressly incorporated into that contract.

2. Our position on generative AI

2.1       We recognise that concerns about generative AI in creative industries are legitimate. Authors and illustrators are entitled to expect that their work will not be appropriated, imitated or commercially exploited without proper rights or permission.

2.2       We also believe Contributors are entitled to protection from unsupported accusations about their own work.

2.3       All rights in works published by the Press are reserved. As stated in the copyright notice in each work, no part of any work may be reproduced, stored in a retrieval system or transmitted in any form or by any means without the prior written permission of Dapple Grey Press Ltd. For the avoidance of doubt, the Press does not permit any work, or any part of it, to be copied into, used to train, fine-tune or test, or otherwise ingested by any generative AI or machine-learning system, or used for text and data mining, without that prior written permission. Nothing in this clause restricts a reader's use of assistive technology for their own personal access to a work.

3. Provenance and rights checks

3.1       Before acquiring a work, we take reasonable steps, appropriate to the work in question, to establish its provenance and that the Contributor holds the necessary rights. Depending on the work, this may include reviewing drafts, layered or working files, sketches, notes and other development materials.

3.2       Our checks rely on human review and professional judgement. We do not consider the output of any automated "AI detection" tool to be reliable evidence of how a work was created, because such tools produce both false positives and false negatives.

3.3       We do not claim that our processes are infallible, and nothing in this policy is a warranty or guarantee about the method by which any particular work was created. Readers and customers are free to decide for themselves whether to place trust in our processes.

4. Confidentiality of working materials

4.1       Drafts, working files, source materials and similar items supplied to us by Contributors are confidential and proprietary to them (or to the Press, where applicable). They are used only for the purposes of acquisition, editing, rights verification and publication.

4.2       Access to such materials is restricted to the Press's acquisitions and editorial function and is not disclosed to third parties, except where required by law, by a court or regulator, or with the Contributor's written consent.

4.3       We will not require any Contributor to publicly disclose or publish their working materials in order to "prove" the human origin of their work in response to public allegations.

5. Concerns about authorship or rights

5.1       If a credible and reasoned concern about the authorship, provenance or rights position of a work is brought to our attention, we will consider it, and where appropriate investigate it privately, fairly and in confidence, with the Contributor given an opportunity to respond.

5.2       Concerns should be sent in writing to dapplegreypress@outlook.com with any supporting evidence. We may not be able to share the outcome of an investigation, for reasons of confidentiality.

5.3       We do not regard the following, alone or in combination, as credible evidence of a rights or provenance issue:

(a)      the output of automated AI-detection software;

(b)     subjective assessments of style, aesthetics or "look";

(c)      online speculation, rumour or social-media commentary; or

(d)     statements that a work "looks AI-generated" without verifiable supporting evidence.

6. Public statements and social media

6.1       The Press does not participate in public disputes or online campaigns concerning the alleged origins of a work. As a matter of policy we do not respond publicly to allegations of the kind described in clause 5.3.

6.2       Silence in response to such an allegation is not an admission, and should not be read as one.

7. Protection of the Press and its Contributors

7.1       We take seriously statements that may damage the reputation or livelihood of the Press or its Contributors. Where a false statement of fact about a work, a Contributor or the Press is published and has caused, or is likely to cause, serious harm to reputation (including, in the case of the Press, serious financial loss), we reserve the right to take appropriate action. This may include a request for removal or correction, a formal letter before claim, and legal proceedings.

7.2       Nothing in this policy limits any right or remedy available to the Press or any Contributor in law, including under the Defamation Act 2013, or restricts anyone's lawful right to freedom of expression, including honest opinion on matters of public interest.

8. Contributor responsibilities

8.1       Contributors are expected to comply with the rights, warranties and AI-related terms of their agreement with the Press, and to tell us promptly if they become aware of any issue affecting the originality or rights position of their work.

8.2       Contributors must not use generative AI tools to create, or to generate any substantive part of, the text, illustrations or other creative content of any work submitted to or published by the Press. This does not prevent the use of assistive or editing tools (including spelling and grammar checkers, accessibility and dyslexia support tools, and translation or transcription aids) that do not generate the creative content of the work. Where a Contributor is unsure whether a tool falls within this restriction, they should ask the Press before using it.

8.3       Contributors must not input any confidential or unpublished material belonging to the Press or to other Contributors (including manuscripts, artwork, editorial feedback or contract terms) into any generative AI tool or service operated by a third party.

8.4       Contributors must tell the Press promptly if they become aware that any part of their work may have been created in breach of this section, or may infringe the rights of any third party. Any review of such a concern will be conducted privately in accordance with section 5.

9. Review and changes

9.1       This policy is reviewed periodically, given the pace at which AI technology and the law are changing. We may amend it at any time by publishing an updated version on our website.

9.2       This policy is governed by the laws of England and Wales.