Ask AI: "Please Describe a Realistic AI Agent For Family History Writing"

I wrote Ask AI: "What Can An AI Agent Do For a Genealogist and Family Historian?" two days ago and want to follow up with the suggested description of a realistic AI Genealogy Agent for family history writing.

1)  My prompt was:  "Please describe a realistic AI genealogy agent that could take a genealogical sketch of a specific ancestor, ancestor images, an obituary of the ancestor, some other family information (e.g., a paragraph from a book), and more and create a document with a biography with family information and images based on the sketch, add stories ("a day in the life," "a life memoir, "how did it feel?," "what did he experience?, etc.) write a poem or song lyrics, create the song, and create a podcast, slide presentation and a video about the life of the ancestor. Is that possible? If not in 2025, when might it be possible?"

2)  The OpenAI ChatGPT5 response was:

That’s a wonderful and very forward-looking question — and the answer is: we’re closer than most people realize.

Let’s imagine a realistic, near-future AI Genealogy Agent — let’s call it AncestorAI — built for exactly your use case: a genealogist who wants to go far beyond collecting facts and instead bring ancestors to life through story, sound, and visuals.

🧬 AncestorAI: The Genealogist’s Creative Companion

1. Data Intake & Comprehension

AncestorAI would begin by ingesting and understanding your source materials, such as:

• Genealogical sketches or notes (from you or Ancestry/FamilySearch profiles)
• Historical documents (census, wills, land deeds, obituaries, letters)
• Photos or portraits
• Book excerpts, local histories, or county biographies
• Family recollections or timelines

Using multimodal AI (text + image + metadata), it would:

• Extract structured genealogical facts (names, dates, places, relationships)
• Understand narrative tone, emotional clues, and family context
• Identify themes (migration, perseverance, tragedy, service, artistry)
• Recognize images and automatically caption, label, or colorize them

2. Narrative Creation and Expansion

Once your ancestor’s life data is understood, AncestorAI could produce:

• Formal biographies (academic/genealogical format)
• Narrative histories (storytelling, memoir style)
• Historical fiction reconstructions (“A day in the life of Henry Carringer in 1850 Pennsylvania”)
• Emotional essays (“What might he have felt when his wife died in 1848?”)
• Thematic essays (“Faith and migration in the Carringer line”)

Each narrative would include:

• Real historical and geographic context (climate, economy, technology of the time)
• Verified records and citations from your inputs
• Optional imaginative but clearly marked reconstructions or dramatizations

3. Creative Outputs

From that biography, AncestorAI could then generate a range of derivative works:

🎭 Storytelling & Multimedia

• Poem or Song Lyrics: written in a fitting poetic style (e.g., 19th-century folk, ballad, or modern country)
• Music Composition: AI models like Suno, Udio, or Mubert could generate melodies and full instrumental tracks
• Narrated Podcast Episode: a voice actor (synthetic or human-sounding AI) reads the life story with background music, transitions, and historical ambiance
• Slide Presentation: includes photos, timelines, maps, and key facts, ready for PowerPoint, Keynote, or Google Slides
• Short Video Documentary: generated from text, voice, and images—complete with transitions, subtitles, and period music

4. Collaboration and Curation

AncestorAI would act like an editor and collaborator:

• You can approve or refine each stage (fact-check, tone adjust, remove embellishment)
• It could cite sources automatically and produce both narrative and research appendices
• It learns your preferred style (concise, emotional, historically detailed, etc.)
• You can generate multiple perspectives (e.g., “Tell this from Sarah Feather’s viewpoint”)

🚀 Is This Possible in 2025?

• Partially — and rapidly improving.

Here’s where we are right now (late 2025):

So:

• You could do all of this today, but you’d need several coordinated tools.
• By 2027–2028, a single integrated genealogy-aware AI platform could plausibly do all of it in one click.

🕰️ Timeline of Realistic Capability

Year What’s Possible

💡 Example Workflow (Today’s Capabilities)

• Upload your ancestor’s biography, photos, and obituary.
• ChatGPT-5 (or similar) generates a polished biography + short story + poem.
• Suno or Udio turns the poem into a song.
• ElevenLabs or Play.ht voices the biography.
• Pika Labs or Runway uses the story and images to produce a short video.
• Canva or PowerPoint auto-builds a slide show.
• A podcast app compiles narration + background music.
• You review, refine, and publish it as a family history “experience.”

If you like, I can mock up what AncestorAI’s interface would look like — showing the input panel (where you upload the biographical sketch, photos, obituary, etc.) and the output options (biography, song, video, etc.) — almost like a product concept sheet or storyboard.

3)  Well, it looks like I'll have another follow-on post here!

4)  At this point in time, I, and others, can accomplish almost all of the above one task at a time, for free.  My current AI-assisted workflow to create engaging information about a specific ancestor is:

Given a genealogical sketch of an ancestor (like those in my 52 Ancestors/Relatives Biographies page), I open the AI Tool (I use Anthropic Claude Sonnet 4.5 as my ghostwriter for writing consistency):

• Upload my ancestor's genealogical sketch (which might include an obituary or published book excerpt) to my AI  tool and ghostwriter - 10 minutes
• Use an appropriate prompt in Claude to create a biographical summary -- 5 minutes
• Use an appropriate prompt in Claude (and Grok) to create a descriptive and celebrative poem and song lyrics -- 5 minutes
• Use an appropriate prompt in Claude to create one or more short stories -- 5 minutes
• Copy all of the AI provided material above into a word processing document -- 10 minutes
• Review and edit each item for accuracy, completeness and consistency -- 30 minutes (redo the item if more information needs to be added)
• Use selected song lyrics in AI Suno to create the song with appropriate style and genre (Suno creates two variations, and I choose one).  "Publish" the song file, and download the video -- 20 minutes
• I don't use a voice synthesizer yet, but I've thought about it!  None are free at this time.
• Use AI Google NotebookLM to read the sketch and create an Audio Overview (a podcast) and a Video Overview (a slide show); download the files (an M4A file for Audio and an MP4 for Video) -- 20 minutes.
• Use AI Google Notebook LM to do the same for each family story created -- 5 minutes each.
• Write a Genea-Musings blog post to capture the  poem and song lyrics, the song video (an MP4 file), and the audio overview link (Blogger doesn't permit embedding an audio file),  and the video overview (an MP4 file) -- 20 minutes.
• Write a Genea-Musings blog post for each family story created, with an audio overview link and an embedded video overview, including a critical review -- 20 minutes each.
• Write a Genea-Musings blog post to publish the biography, links to the poem, song, audio overview, video overview and family stories as required, including a cirtical review -- 30 minutes.
• Create an AI image of families or settings to have an image in my blog posts using ChatGPT, Gemini, Grok or CoPilot - 10 minutes. 
• Up0load the Video Overview MP4 file directly to YouTube on my Randy Seaver, Geneaholic channel -- 10 minutes.

Realistically it comes out to 3 to 4 hours of time to do all of the tasks for each ancestor.  But I spread the work over a week or two as I have more ideas for stories.  Plus, I work on several ancestors at once as I look ahead to the next ancestor.

All of my AI biographies, family stories, songs and poems, YouTube videos, etc. are on my Randy's Genealogy and AI page on Genea-Musings.

5)  I have now created a biography, poem, song, audio overview and video overview for 61 ancestors - from my parents, grandparents, first, second and third great-grandparents.  I don't have a YouTube video for every ancestor yet. I have written two ABC Biographies each week for the past 30 weeks, plus blog posts for the family stories. I have more family stories to do for many of the ancestors.  

That covers all of my ancestors that lived well into the 19th century, and all of them that lived into the 20th century.  My thinking is to NOT do the 4th great-grandparents because the material I have for the genealogical sketches is thin for many of them - in some cases only birth, marriage, death and burial records, perhaps probate and military records, and no obituaries, family stories, or published excerpts at all.  

In future weeks, I may do my entire Seaver line back to 1634, and my Revolutionary War soldiers. I need to start working on genealogical sketches and family stories for my wife's ancestors too (I may go back through only her 2nd great-grandparents - I have to leave my daughters and grandchildren something to do if they are inspired to do so).

6)  Where am I going with all of this?  My current plan is to write relatively short eBooks in PDF format for each ancestral family (husband and wife biographies merged together), plus the family stories, poem and song lyrics, and links to the song and video overview on YouTube, with the sourced biographical sketch at the end, including some record images.  

I could disseminate them over time to my close family members (brothers, daughters, grandchildren, cousins) in a monthly RandyGram email.  Maybe one or two of them would collect them for posterity. I will also put them in my FOREVER account and hope that my daughters and gtrandchildren can find them. More work, but distributing my work to the family is important to me. My plans may change as the technology improves.

I have 4 to 8 hours available every day to do genealogy work (researching, DNA matches, writing, blogging, YouTubing, webinars, and society work) because I am retired, have few close friends nearby, try to be frugal, and try to keep my mind stimulated.  I have a lot of fun, and the AI-generated  story ideas and results really touch my heart.  I truly appreciate all of the people who contributed to my DNA.  I feel mortal and hope to postpone the inevitable as long as I can by staying busy and sharing my genealogy fun.

7)  Can an AI Agent or Agentic AI do this better, quicker and more accurately?  Someone more proficient than me needs to write the AI  Agent and make it work.  Right now, I don't think it can, but in a year or two we may have useful AI Agents for genealogy research and family history writing. 

==============================================

Links to my blog posts about using Artificial Intelligence are on my Randy's AI and Genealogy page.  Links to AI information and articles about Artificial Intelligence in Genealogy by other genealogists are on my AI and Genealogy Compendium page.

Copyright (c) 2025, Randall J. Seaver

The URL for this post is:  

Please comment on this post on the website by clicking the URL above and then the "Comments" link at the bottom of each post.  Share it on Twitter, Facebook, or Pinterest using the icons below.  Or contact me by email at randy.seaver@gmail.com.  Note that all comments are moderated, so they may not appear immediately.

Subscribe to receive a free daily email from Genea-Musings using www.Blogtrottr.com.