AI Tool Policy

Governing Principle

Authors are personally responsible for every proposition in their manuscript. Each statement of law, fact, or data must be verified by the author against the source actually consulted, whether or not an AI tool was used in the writing process.

Authorship

Generative AI systems cannot be listed as authors or co-authors, because they cannot take responsibility for the work, respond to reviewers, or approve the final version.

Tiered Disclosure

Authors must declare their use of AI tools in the submission form and in a statement at the end of the manuscript, using one of four tiers:

Tier

Use

Required statement

Tier 0

No AI tools used

“No generative AI tools were used in preparing this manuscript.”

Tier 1

Grammar, style, or translation only

Name of the tool and the purpose of use

Tier 2

Literature searching, summarising, data coding, or transcription assistance

Name of the tool, the tasks performed, and how outputs were verified

Tier 3

AI-generated text retained in the scholarly content

Full disclosure of the tool, the passages concerned, and the verification carried out; subject to editorial approval

Prohibited Uses

The following uses are prohibited and constitute misconduct:

1.        generating or paraphrasing statutory, regulatory, or judicial text instead of quoting the authentic source;

2.        generating citations, references, or case names that the author has not personally verified;

3.        fabricating or altering field data, interview material, maps, or land records;

4.        generating images, maps, or figures presented as documentary evidence;

5.        using AI to disguise plagiarism or duplicate publication; and

6.        producing peer review reports or editorial decisions with AI.

Reviewers and Editors

Reviewers must not upload any part of a manuscript under review into an AI system, as doing so breaches confidentiality. Editors may use AI tools only for administrative tasks, such as reference checking or formatting, and every result must be confirmed by a human editor. Editorial decisions are never delegated to AI.

Detection

DAB does not rely on AI-detection software to make decisions, since such tools are unreliable. The journal instead relies on author declarations, source verification, and reviewer scrutiny. DAB welcomes the responsible use of language tools by authors whose first language is not English.