AI for Genealogy Research: Find Leads and Verify the Evidence

AI helps genealogy researchers develop search strategies, work through transcription problems, and evaluate what a record might indicate. It cannot verify primary sources, access subscription genealogy databases on your behalf, or replace the standard of evidence that distinguishes genealogical proof from family legend.

The most useful AI sessions in genealogy are the ones that help you plan the next search rather than the ones that give you an answer. A tool like fenbs keeps your verified source links, research decisions, and the current state of each research question in one place that carries forward across AI sessions.

What AI does well in genealogy research

Genealogy research involves working with historical records, unfamiliar naming conventions, handwritten documents, and fragmentary evidence. AI assists with several of the challenges this creates.

• Suggesting search strategies for a specific time period, location, or record type.
• Explaining historical context: what records existed in a given place and time, what events affected record keeping.
• Helping transcribe or interpret difficult handwriting when given a description of the characters.
• Identifying alternative spellings of surnames or given names common in specific regions or immigration patterns.
• Explaining the difference between record types: what a death certificate can and cannot tell you versus a death record in a church register.
• Suggesting DNA match interpretation approaches and which tools are relevant for specific scenarios.
• Drafting research reports or correspondence to archives and record repositories.

AI cannot search genealogy databases for you, access records behind subscription paywalls, or verify that a document actually exists. The research itself still happens in Ancestry, FamilySearch, FindMyPast, the National Archives, or the relevant state or county records office.

Comparing AI tools for genealogy research

Different AI tools bring different strengths to genealogy work. Check current features on vendor pages before relying on any specific capability.

FamilySearch is a strong starting point for most US genealogy because the collection is extensive and free to access. For more recent records, US federal census records through 1950 are available through Ancestry, FamilySearch, and MyHeritage, each with different indexing and transcription quality.

Separating evidence from inference

The Genealogical Proof Standard, used by the Board for Certification of Genealogists (BCG), requires a reasonably exhaustive search, complete and accurate citations, analysis and correlation of evidence, resolution of conflicting evidence, and a soundly reasoned conclusion. AI can assist with the search and the reasoning; the standard itself is yours to meet.

The most common genealogy error AI can introduce is treating inference as evidence. If an AI suggests "this is probably the same person based on the name and approximate dates," that is a lead to investigate, not a conclusion to record as fact. Write down what the record actually says, and keep your interpretation separate.

Ask AI to identify the weaknesses in a proposed connection: what evidence would be needed to confirm it, what evidence would disprove it, and what alternative explanations exist. This kind of adversarial reasoning is where AI adds real value to genealogical analysis.

DNA research and AI

AI tools can explain DNA match interpretation approaches, help you think through shared centimorgan ranges and their relationship probability implications, and suggest strategies for working with large match lists. For specific interpretations, the tools built into the testing company's platform (such as AncestryDNA ThruLines or 23andMe DNA Relatives) combine the analysis with the actual match data you cannot give to a general AI tool.

Do not give general-purpose AI tools identifying information about DNA matches or other living people. Your genealogical data includes sensitive information about relatives who have not consented to having their data analyzed by AI tools.

Research decisions worth recording

Genealogy research produces decisions that matter for future work: which of two people with the same name you have determined is your ancestor, what evidence led to that conclusion, and what would change your mind. These decisions can be revisited years later as new records are digitized and released.

Recording the reasoning behind a research conclusion, not just the conclusion, is what lets you return to it honestly. fenbs Decisions and rules works well for this: the research question, the evidence considered, the conclusion reached, and what additional evidence would confirm or overturn it.

How to do this with fenbs

1. Create a free account. Add a genealogy research project. Create one task per research question: finding a birth record, identifying parents of a specific ancestor, confirming an immigration date.
2. Record verified sources in Links and listings as you find them: archive name, record type, document identifier, and what it says. These become your evidence base for each research question.
3. Record research decisions in Decisions and rules: which person you have identified as your ancestor and on what evidence, which conflicting evidence you have resolved and how.
4. When an AI research suggestion leads nowhere or turns out to be wrong, add a Lesson: the suggestion, the result, and what a better search strategy would be. This prevents retracing dead ends.
5. Connect your AI app from Integrations and ask it to read your research project context before each session. It starts from your current evidence base and open questions.

For AI tools used across research workflows, see Google AI tools and how to use AI at work. The what is an AI brain post explains the broader concept of keeping research context outside any one AI session.