Disclosing AI Assistance in Genealogy Writing: A How-To Guide

Disclosing AI Assistance in Genealogy Writing: A How-To Guide

Disclosing AI Assistance in Genealogy Writing: Artificial intelligence tools have become increasingly accessible to genealogists over the past eighteen months. Large language models can draft correspondence, summarize research findings, and suggest narrative structures. Image recognition tools can extract text from documents. Translation algorithms can render foreign-language records into English within seconds. These capabilities offer genuine efficiency gains, yet they also introduce questions about attribution, methodology, and professional transparency.

Introduction: The Transparency Imperative

Genealogical publication has always rested on principles of reproducibility and verification. Readers must be able to trace an author’s reasoning from sources through analysis to conclusion. When AI tools participate in any stage of this process—whether through research assistance, content generation, or editorial refinement—the question emerges: what should authors disclose, and when?

This article examines current professional standards as they apply to AI use in genealogical writing. It identifies where existing ethics codes and publication guidelines provide clear direction, where interpretive gaps remain, and what disclosure practices genealogists might adopt to maintain professional credibility during this period of technological integration.

Defining AI Use in Genealogical Writing

Not all computational assistance constitutes “AI use” in the sense requiring disclosure consideration. Understanding the distinction between different tool categories helps genealogists make informed decisions about transparency.

• Spell-checking and grammar correction. Tools that flag typographical errors, suggest punctuation corrections, or identify subject-verb disagreement operate at the mechanical level. These functions have been standard in word processing software for decades. Most professional contexts treat spell-checkers as analogous to dictionaries—reference tools that support but do not substitute for authorial judgment. Current genealogical standards do not typically require disclosure of spelling or grammar-checking tools.

• Translation assistance. AI-powered translation tools can convert foreign-language documents into English with increasing accuracy. When genealogists use these tools to understand source material, they engage in a form of analytical assistance. The translated text becomes an intermediate step in research rather than published content itself. However, when translated passages appear in published work, citation practices vary. Some genealogists note the translation tool used; others treat machine translation as they would dictionary-assisted translation, disclosing only when the translation itself required interpretive judgment that might affect conclusions.

• Content generation. Large language models can draft entire paragraphs, generate narrative structures, or compose prose based on prompts. This category represents a qualitative difference from mechanical correction tools. When AI generates sentences, paragraphs, or structural frameworks that appear in published work—even if subsequently edited—the question of authorship becomes material. Professional standards increasingly treat undisclosed AI-generated content as inconsistent with claims of sole authorship.

• Research assistance and analysis. Some AI tools can summarize documents, extract data from images, suggest source correlations, or identify patterns in genealogical evidence. These applications blur boundaries between research tool and analytical partner. When AI systems contribute to the interpretive process—suggesting connections between records, proposing identity resolutions, or summarizing evidentiary patterns—they enter territory traditionally associated with genealogical reasoning itself.

The distinction that matters professionally is whether AI use affects the reproducibility of published conclusions or the attribution of intellectual work. Mechanical corrections rarely affect either. Content generation and analytical assistance often affect both.

Disclosure Matters! Here’s Why

Transparency about methodology has long been central to genealogical publishing. The Genealogical Proof Standard requires that published conclusions rest on “reasonably exhaustive research,” “complete and accurate source citations,” “thorough analysis and correlation,” and “resolution of conflicting evidence.” Readers must be able to evaluate whether these standards have been met.

• Reproducibility and verification. When researchers use AI tools that affect their analytical process or written conclusions, readers benefit from knowing this context. If an AI tool suggested a connection between two records, reviewers can consider whether that suggestion rested on pattern recognition humans might have achieved independently, or whether it reflected computational capabilities beyond human replication. If AI tools generated portions of narrative text, readers can assess whether the prose accurately reflects the author’s research findings or whether generation artifacts might have introduced subtle distortions.

• Professional credibility. Genealogical credentials rest partly on demonstrated competence in research, analysis, and writing. When authors present work as their own, professional readers assume the intellectual labor—including the reasoning, synthesis, and composition—originated with the credited author. Undisclosed AI generation can create misattribution, particularly when the AI contribution was substantial. This concern has intensified as certification bodies have clarified that portfolios submitted for professional evaluation must represent applicants’ own work.

• Intellectual honesty. Academic and professional publishing conventions have historically required authors to acknowledge assistance that substantially contributed to published work. Research assistants receive acknowledgment. Editors who substantially reshape prose receive credit. The principle extends logically to AI tools when their contribution exceeds mechanical correction. Disclosure becomes an extension of existing attribution norms rather than a new ethical requirement.

• Reader expectations. Audiences increasingly recognize that AI tools vary widely in capability and application. Readers who learn an author used AI spell-checking typically do not reconsider the work’s credibility. Readers who discover an author used AI to generate substantial portions of text without disclosure may question whether other undisclosed shortcuts affected research quality or analytical rigor. Transparent disclosure allows readers to calibrate their trust appropriately.

Current Standards and the Gap

Major genealogical organizations maintain ethics codes and publication standards, yet as of January 2026, most do not include explicit AI disclosure requirements in their general guidance for published work. This reflects the pace of technological change rather than indifference to transparency principles.

• The Board for Certification of Genealogists. BCG has provided the most explicit guidance to date, though it applies specifically to certification portfolios rather than to all published work by certified genealogists. In July 2024, BCG issued an interpretation stating that work submitted in new CG® application portfolios must be entirely the applicant’s own and may not be written or generated by AI. The interpretation permits applicants to use AI-powered search tools, spelling and grammar checking tools, and translation tools (with the exception of the BCG-Supplied Document). AI may not be used for any aspect of the BCG-Supplied Document Work requirement, including transcribing, translating, abstracting, analyzing, or research planning. This guidance took effect with BCG’s July 1, 2025, Application Guide and establishes a clear boundary for certification applicants: AI-generated content is incompatible with demonstrating personal competence in genealogical writing.

BCG’s position clarifies expectations for those seeking certification but does not explicitly address disclosure requirements for board-certified genealogists’ subsequent publications. The implication, consistent with BCG’s emphasis on competence and ethical transparency, is that certified genealogists should apply similar principles to published work—distinguishing mechanical tools from generative assistance and maintaining honest attribution.

• The Association of Professional Genealogists. APG’s Code of Ethics requires that genealogists “accurately represent their credentials, competence, knowledge, and scope of work” and “inform clients when they are unable to follow professionally accepted standards or answer research questions.” While these provisions clearly apply to client work, they do not yet explicitly address AI disclosure in publications. APG has devoted increasing attention to AI integration through educational programming, including sessions at the 2025 Professional Management Conference and the 2026 Symposium, where “Ethical and Legal Considerations: AI, Copyright, DNA” was featured. These discussions suggest evolving organizational awareness, though formal policy guidance for AI disclosure in published work remains under development as of early 2026.

• The National Genealogical Society. NGS has emphasized education about AI capabilities and limitations through partnership programming, including the “Navigating the AI Frontier” series developed with AI Program Director Steve Little. NGS’s Standards for Sharing Information with Others calls for genealogists to “respect the traditions of the genealogical community” and “be sensitive to the needs of fellow genealogists.” These principles support transparency but do not mandate specific AI disclosure practices. As of January 2026, NGS has not issued explicit AI disclosure requirements for articles submitted to the National Genealogical Society Quarterly or other NGS publications, though editors may request clarification about AI use on a case-by-case basis.

• The Genealogical Proof Standard. The GPS itself predates widespread AI availability and does not reference artificial intelligence explicitly. However, its requirements for thorough analysis, complete citations, and resolution of conflicting evidence all rest on assumptions about transparent methodology. When AI tools contribute to analysis or writing, the question becomes whether undisclosed use affects readers’ ability to evaluate whether GPS standards have been met. The answer likely depends on the extent and nature of AI involvement.

• Academic and professional publishing. Genealogists who publish in academic journals or with scholarly presses encounter AI policies established by those venues. Major academic publishers including Elsevier, Springer Nature, Taylor & Francis, and SAGE have implemented AI disclosure requirements as of 2024-2025. These policies typically distinguish between assistive tools (language editing, grammar checking) and generative tools (content creation, data analysis). Most require authors to disclose generative AI use and describe its application. Genealogists submitting to these venues must comply with venue-specific requirements regardless of genealogy-specific standards.

The gap between general professional ethics and AI-specific guidance creates an interim period during which genealogists must exercise professional judgment. The principles underlying existing standards—attribution, reproducibility, honest representation of competence—provide direction even where explicit AI policies have not yet been codified.

Publisher and Self-Publishing Requirements

If planning to publish a genealogy book (whether through a traditional publisher or via self-publishing platforms like Amazon or Lulu), there are two angles to consider: legal/contractual requirements and ethical reader transparency. Both are important in the context of AI.

Amazon KDP

Amazon’s Kindle Direct Publishing (KDP) platform is a go-to choice for self-publishing family histories and genealogical reference books. In late 2023, facing a surge of AI-generated books, Amazon KDP implemented a new rule: authors must inform Amazon if a book’s content (text, images, or translations) was generated by AI. This policy, announced in September 2023, was designed to improve transparency and help Amazon manage the influx of AI-created material.

When uploading a manuscript on KDP now, Amazon asks authors to declare whether the content is “AI-generated”. According to KDP’s guidelines, “AI-generated” means content that was created by an AI tool, even if an author heavily edited it afterward. E.g., if using ChatGPT to write whole paragraphs of a book, that counts as AI-generated. By contrast, “AI-assisted” content – content written by an author, with minor help from AI for editing or idea brainstorming – does not have to be disclosed. Using an AI spell-checker or asking ChatGPT for outline ideas wouldn’t require ticking the disclosure box; using it to draft chapters would.

It’s crucial to answer Amazon’s disclosure question honestly. Amazon doesn’t (at this time) publicly label the book as AI-generated, but internally they track it. Not disclosing properly could risk account termination if later discovered. The Authors Guild welcomed this Amazon policy as a “first step” for transparency. It helps protect readers from unknowingly buying books that might be fully AI-produced and potentially less reliable.

That covers informing Amazon. Separately, should an author tell readers? Amazon doesn’t require authors to place an AI disclosure in the book itself, but they might choose to. Ethically, many authors do include a note on the copyright page or introduction if AI played a significant role. Aside from honesty, there’s a practical reason: purely AI-created text as a whole and as generated cannot be copyrighted under U.S. law (because it lacks human authorship). If a book is largely AI-written, technically some content might not be protected by copyright. Disclosing AI use and clarifying that text was edited and taking responsibility can help authors assert their claim over the final work. For instance, an author’s note might read:

“Portions of the initial draft were generated by AI. The author has substantially revised this material and affirms the originality of the final content.”

This both informs the reader and stakes the author’s position as the creative mind overseeing the AI output.

Lulu and Other Publishers

Other self-publishing platforms like Lulu.com have begun to address AI as well. While Lulu’s terms of service as of this writing don’t mandate an AI disclosure to them, they, like all publishers, prohibit plagiarism and misleading authorship. That effectively means all AI-generated content must be handled carefully. This was aimed at stopping low-quality, mass-produced AI books. If an author is publishing a single family genealogy, they are most likely not the target of those rules, but they should still ensure that AI-assisted content is original and not copied from some source the AI platform may have referenced.

Traditional publishing houses (for example, if an author submits a manuscript to a genealogy or history publisher) currently treat AI use on a case-by-case basis. Many publishers’ contracts require an author to warrant that the work is original and that they are the author. If using AI to write large parts, there could be gray area – did the author create it, or the AI? To be safe, inform the publisher during the process if AI was used. They may ask the author to remove or rewrite those parts due to copyright concerns or to maintain quality. Some publishers might be fine with it as long as the author (the human), edited and takes responsibility (similar to the academic journal world where AI can assist but not be listed as an author). Others might insist on an explicit acknowledgment in the book.

On self-publishing platforms, the author is essentially the publisher, so the onus is on them to be ethical. Beyond satisfying Amazon’s and Lulu’s rules, consider the readers. Genealogy readers often expect a note on research methods – that’s a great place to mention AI. For example, in a preface an author might write:

“In writing this family history, I experimented with an AI tool to generate early drafts of some historical descriptions. Every AI-generated passage was verified against sources and rewritten for accuracy.”

This kind of transparency can actually enhance credibility, showing that the author took innovative approaches but also exercised due diligence.

Key takeaway: Major outlets like Amazon KDP require disclosing AI-generated content to the platform, and others are moving in that direction. Always read the latest content guidelines of a chosen publisher. And regardless of platform rules, include a disclosure for the reader’s sake if AI had a hand in the text. It’s both an ethical best practice and a shield against potential legal pitfalls.

Disclosure Best Practices

In the absence of universal mandatory requirements, best practices emerge from professional principles, emerging organizational guidance, and practical considerations about reader trust. Genealogists can approach disclosure decisions by considering the nature and extent of AI involvement.

• When disclosure is expected. Professional standards increasingly converge on the expectation that genealogists disclose AI use when AI tools generated substantial portions of published text or when AI tools contributed materially to analytical conclusions. “Substantial” typically means more than isolated sentence corrections but less than minimal involvement. A working threshold: if the AI tool produced paragraphs of prose that the author then edited, disclosure allows readers to understand the composition process. If the AI tool suggested connections between records that became central to the article’s argument, disclosure helps readers evaluate the reasoning pathway.

Disclosure is particularly expected when submitting work to venues with explicit AI policies. Journal editors, conference organizers, and book publishers increasingly include AI disclosure questions in submission guidelines. Genealogists should review submission requirements carefully and respond accurately.

For work submitted for professional evaluation—including certification applications, award submissions, or peer review—disclosure standards are higher. BCG’s position that certification portfolios must be the applicant’s own work, without AI generation, reflects a principle applicable to other evaluation contexts. If evaluation criteria include demonstrated writing competence, AI-generated prose may be incompatible with evaluation purposes even if it would be acceptable in other publication contexts.

• When disclosure may be optional. Genealogists need not typically disclose use of spell-checking tools, basic grammar checkers, or formatting assistance. These tools function at the mechanical rather than compositional level. Similarly, using AI-powered search tools—such as genealogy platform hints, search engine algorithms, or database query optimization—typically does not require disclosure unless the specific tool’s methodology affects research conclusions in ways readers should understand.

Translation tools occupy a middle ground. When genealogists use AI translation to understand source material but do not publish the translations directly, disclosure may be optional. When translated text appears in published work, some genealogists note the translation method in citations while others do not. Best practice likely aligns with the extent to which translation choices affected interpretation. If the AI translation introduced ambiguities or alternative readings that influenced conclusions, disclosure supports reproducibility. If translation was straightforward and confirmable through other methods, disclosure becomes less material.

• When disclosure is unnecessary. Using general-purpose software tools that happen to incorporate AI features does not necessarily require disclosure. Many word processors, email clients, and productivity applications now include AI-powered features such as autocomplete suggestions or formatting assistance. When these features operate transparently in the background and do not affect substantive content, disclosure would create more confusion than clarity. The principle remains: disclose when AI involvement affects attribution or reproducibility, not merely because AI algorithms operated somewhere in the technological chain.

• Decision framework. Genealogists uncertain about whether to disclose AI use might consider three questions. First, did the AI tool contribute to the intellectual content of the publication, or did it provide mechanical assistance? Second, would readers evaluating the work’s quality or credibility benefit from knowing about the AI tool’s involvement? Third, does the publication venue have specific disclosure requirements? An affirmative answer to any question suggests disclosure would be appropriate.

Sample Disclosure Language

Effective disclosure statements are concise, specific, and contextually appropriate. They inform readers without suggesting apology or defensiveness. The following examples illustrate disclosure approaches for different publication contexts.

• For peer-reviewed journal articles. Many genealogical and historical journals now request author statements about AI use, either during submission or in published articles. Appropriate disclosure language might read:“The author used [specific AI tool name] to assist with [specific function, e.g., ‘language editing and grammar checking’ or ‘generating an initial draft outline that was subsequently revised’]. The author reviewed and revised all content and takes full responsibility for the article’s accuracy and interpretation.”

When AI contributed to research analysis rather than writing, the disclosure might specify: “The author used [tool name] to identify potential record linkages, which were subsequently verified through manual examination of original sources. All analytical conclusions represent the author’s independent judgment.”

• For books and monographs. Book acknowledgments sections traditionally credit individuals and institutions that contributed to the work. AI disclosure can follow this convention:“I used [AI tool name] for [specific purpose] during the preparation of this manuscript. All narrative content, analysis, and conclusions represent my own research and reasoning.”

Some authors prefer transparency about process in prefaces or methodology sections:

“Research for this book was conducted between [dates]. Translation of non-English documents was assisted by [tool name], though all translations were reviewed for accuracy and alternative interpretations. No AI tools were used for content generation or analytical reasoning.”

• For blog posts and informal publications. Genealogy blogs and informal online writing allow more conversational disclosure. Authors might note:“I experimented with [AI tool] to see how it might streamline the writing process for this post. I used it to generate an initial draft of the opening section, which I then rewrote to better reflect my research findings. Everything else is written in my usual way.”

Brief, good-faith disclosure often suffices:

“This post was written with grammar-checking assistance from [tool name].” or “I used [AI tool] to help organize my thoughts before writing. The words are all mine.”

• For social media. Short-form platforms like Twitter or Instagram typically do not accommodate detailed methodological disclosure. When AI use is material to shared content, authors might include brief notes:“Summary generated with AI assistance from my research notes” or “Draft caption by [AI tool], edited by me.”When AI use was minimal or mechanical, no disclosure is typically necessary.
• For client reports and professional work products. Genealogists producing work for clients should consider including AI disclosure in methodology sections when relevant:“Research was conducted using standard genealogical databases and archives. Document transcription was assisted by [OCR tool name]. All analysis and conclusions represent my professional judgment based on the evidence gathered.”

Conclusion: Transparency as Professional Strength

The genealogical community’s response to AI integration will shape professional credibility for years to come. Transparency about methodology, including AI tool use, reinforces rather than undermines professional authority. Readers who understand how authors work—what tools they employ, what assistance they receive, where their own expertise and judgment operate—can better evaluate published genealogical conclusions.

Disclosure practices need not be burdensome. Brief, honest statements about material AI contributions suffice to maintain trust. Genealogists who disclose appropriately demonstrate respect for professional standards, reader expectations, and the reproducibility principles that underlie sound genealogical work.

As organizational guidance continues to evolve, genealogists can look to established principles for direction. The emphasis on honest attribution, transparent methodology, and demonstrated competence that has always characterized professional genealogy applies equally to AI-assisted work. Tools change, but the commitment to trustworthy, verifiable research endures.

Professional genealogists navigating AI disclosure decisions serve their readers, their profession, and their own credibility by erring toward transparency. When uncertain whether to disclose, choosing disclosure rarely introduces risk. Omitting material information about AI contributions, by contrast, can compromise professional standing if later discovered. The emerging norm appears clear: disclose when disclosure serves reader understanding, reproducibility, or honest attribution. This approach honors genealogical traditions while embracing the practical realities of contemporary research and writing.

* * *

Author’s Note: I want to be transparent that this content – AI and Genealogy Publication Disclosure – was created in part with the help of an artificial intelligence (AI) language model – Claude Sonnet 4.5. The AI assisted in generating an early draft of the content, but every paragraph was subsequently reviewed, edited, and refined by me. The final content is the result of extensive human curation and creativity. I am proud to present this work and assure readers that while AI was a tool in the process, the story, style, and substance have been carefully shaped by the author.