Friday, November 20, 2026

When Your 22nd Great-Grandfather Becomes Your Cousin: An Ancestry Relationship Mystery

 

How much can genealogists trust the relationship labels in their online family trees?

I've spent considerable time reviewing the medieval pedigrees for my multivolume family history, Echoes of Britannia. The process involves checking parent-child relationships, comparing dates, identifying duplicate individuals, and correcting ancestral lines that have become tangled over centuries.

Recently, I discovered something peculiar in my Ancestry.com family tree.

Thomas Grosvenor (1377–1429) appeared as my 21st great-grandfather. His father, Robert Grosvenor (about 1342–1396), however, was identified as my second cousin 19 times removed.

Wait. What? If Thomas is my 21st great-grandfather and Robert is his father, Robert should be my 22nd great-grandfather.

The two men were correctly connected in my tree. So why was Ancestry identifying one as a direct ancestor and the other as a distant cousin?

It gets stranger! My husband and I share ancestral families, although we descend from some of them through different lines. I wondered whether Ancestry was selecting an alternative relationship path rather than following the direct ancestral line.

I changed the home person in the tree from myself to my husband.

The results were even more convoluted.

Thomas Grosvenor was now identified as the husband of my husband's 10th cousin 16 times removed.

His father, Robert, became the husband of the mother-in-law of my husband's 4th cousin 20 times removed.

Try putting that on a pedigree chart!

The relationships weren't merely different. Ancestry was now describing connections through marriage rather than displaying the father-son relationship as part of a single ancestral line.

I decided to do one more experiment

I decided to make Thomas Grosvenor the home person.

This time, Thomas had no relationship label, as expected. More importantly, Robert Grosvenor appeared simply as father.

That was correct. The software recognized the immediate parent-child connection. The problem appeared when Ancestry calculated relationships across many generations.

I had not changed the parent-child link between Thomas and Robert. Yet depending on the home person, Ancestry displayed dramatically different relationships.

So, why does this matter? Genealogists working with medieval ancestry frequently encounter pedigree collapse. The same individuals may appear in multiple ancestral lines because distant relatives married one another over generations.

A person might be a direct ancestor through one line, a distant cousin through another, and connected by marriage through still another.

That is not unusual.

What concerns me is that Ancestry displays a single relationship label without making it clear, in these screenshots, that other relationships may exist.

When a correctly linked father and son are identified as a great-grandfather and a distant cousin, a researcher might reasonably conclude that something is wrong with the tree.

Worse, someone attempting to correct the apparent discrepancy might change an otherwise valid genealogical connection.

That is a genuine research concern. In my first 5 volumes, I've noticed 92 wrong relationships.

Is This a Software Glitch?

I cannot determine from these examples whether Ancestry's relationship calculator contains a programming error or is simply choosing one relationship path according to rules that aren't apparent to the user.

But I can demonstrate that the displayed labels do not consistently follow the direct ancestral path.

And that distinction matters.

The father-son relationship remained intact. It was the description of each man's relationship to the selected home person that changed.

For researchers who use Ancestry's labels to help identify direct ancestors, this behavior can be misleading.

A Warning for Researchers

My experience suggests several precautions:

  • Don't assume Ancestry's displayed relationship is the only relationship you have with an individual.
  • Verify direct ancestral lines generation by generation.
  • Be particularly cautious when working with interconnected families or pedigree collapse.
  • Don't alter documented parent-child relationships simply to make Ancestry's relationship label appear correct.
  • Remember that a software-generated relationship is not genealogical evidence.

For Echoes of Britannia, I will continue relying on documented relationships and carefully constructed pedigree charts rather than Ancestry's automatically generated labels.

But I would like Ancestry to address this issue.

When multiple relationships exist, how does Ancestry determine which one to display? And why isn't the direct ancestral relationship given priority when one exists?

Those seem like reasonable questions for a product used by millions of family historians. Especially one that costs a lot each year AND KEEPS WANTING MORE OF OUR MONEY FOR PRO TOOLS WHICH IMHO, SHOULD BE PART OF THE PACKET TO INSURE INFORMATION IS CORRECT.

After all, genealogy is complicated enough without software turning your great-grandfather into your cousin.

Have you encountered similar relationship discrepancies in your Ancestry family tree? I'd be interested in hearing what you've discovered.

Tuesday, November 17, 2026

From Indentured Servant to Research Assistant: My Journey with AI

 

AI Image

When people ask how I use AI in genealogy, they often assume I simply type a question and accept whatever answer appears on the screen.

Nothing could be further from the truth. Our relationship didn't begin with trust. It began with skepticism.

At first, AI was my indentured servant.

I gave it small, well-defined tasks and watched closely. Summarize this article. Explain this record. Rewrite this paragraph. Find an error. Every response was checked. Every citation was verified. Every conclusion was questioned.

Sometimes it got things wrong. Sometimes it made assumptions. Sometimes it confidently presented information that simply wasn't supported by the evidence.

As genealogists, we have a name for that. Bad research! So I tested it again.

And again. Eventually, something changed.

I realized that if I clearly explained my expectations, required citations, and insisted that every statement be supported by evidence, the quality improved dramatically. AI wasn't learning about genealogy in the human sense, but I was learning how to work with it effectively.

That was the apprentice stage.

Instead of assigning isolated tasks, I began involving AI in larger projects. It helped organize research logs, suggest outlines, identify inconsistencies, and point out places where my reasoning could be strengthened. It wasn't replacing my judgment. It was helping me think more systematically.

Today, I think of AI as a research assistant.

Notice what I didn't say. I didn't say expert. I didn't say genealogist. And I certainly didn't say author. Those jobs are still mine.

A good research assistant doesn't decide whether the evidence proves a relationship. It doesn't determine whether conflicting records can be reconciled. It doesn't know when a family story should be believed or discarded.

Those decisions require experience, historical context, and professional judgment.

What AI does remarkably well is something different. It organizes information, checks for consistency, synthesizes large amounts of material, and suggests alternative ways of looking at a problem. It catches details I might overlook after staring at the same document for hours.

Most importantly, it gives me someone to argue with.

That may sound odd, but some of my best discoveries have come from asking AI to challenge my assumptions rather than agree with them. If my conclusion is sound, it survives the questioning. If it isn't, I'd rather discover the weakness before publishing than after.

Looking back, I realize something else changed. In the beginning, I gave AI individual tasks. Today, I delegate entire processes.

Instead of asking it to summarize one article, I ask it to help design a workshop, organize a book, review the consistency of a manuscript, or identify gaps in a research plan. Those are collaborative efforts rather than isolated assignments.

That shift didn't happen overnight. It was earned through months of testing, refining, correcting, and learning from each other. Ironically, the biggest lesson wasn't about artificial intelligence at all. It was about partnership.

The best research assistants don't replace researchers. They make researchers better. Perhaps that's the real promise of AI for genealogy.

Not replacing the genealogist, but giving each of us another thoughtful set of eyes as we pursue the stories of those who came before us.

Saturday, November 14, 2026

Every Family Has a Story. Not Everyone Has to Be a Writer.

AI Image

Our ancestors did not all leave beautifully written memoirs. Some left letters filled with spelling mistakes. Others left little more than a signature, a mark on a document, or a story passed down through generations. We treasure those fragments today because they offer glimpses into lives that might otherwise have been forgotten. Why, then, would we discourage someone from using dictation software or AI to preserve their family's history? The words may need polishing, but the memories, experiences, and relationships belong to the people who lived them. Our responsibility is to preserve those stories faithfully, not to judge the tools that helped bring them to the page.

I know there are those who disagree with me as I was recently chastized at a conference after my presentation for recommending those who are uncomfortable with writing to use AI. The attendee told me I was ignorant several times and that she was fired as a high school English teacher due to AI. Having spent 44 years in education, I seriously doubt that was the cause of the termination. Personally, I applaud whatever school district got this woman out of the classroom. It was difficult enough as a mature adult to have to handle her offensive attitude; I can't imagine how a teen would have felt being subjected to her.

Yet, I learned something from this encounter. The attendee told me she stopped listening to me when I mentioned AI. That was about Slide 3. She lost all of the other valuable information that may have helped her with her family history because she closed her mind. That's sad. A closed mindset blocks new ideas, resists feedback and prevents us from learning the truth. Sure, the truth can hurt sometimes but it helps us prevent falling for the same mistake repeatedly and avoiding harm.

If you are finding yourself stuck in a rut of your beliefs try this:

Pause next time you hear something that feels uncomfortable or weird. Ask yourself - What if they are onto something here?

Before automatically making a judgement, ask yourself - What might I be missing?

Consider the other person's viewpoint - think of 3 logical reasons why they may be right

Intentionally seek out new ways to do something you always do. Read a book from an opposing view, LISTEN instead of speak and importantly, try to stop having to be right all the time - that's the enemy of learning.

These small steps might just help you solve that brickwall, get a persnickety relative to share info with you, and brighten your outlook. It can't hurt!

Wednesday, November 11, 2026

Why Your Family Tree Will Never Be 100% Accurate

 

https://humanlineage.org/en

Genealogists strive for accuracy. We search for records, evaluate evidence, and correct errors. But I've come to believe that no family tree will ever be 100% accurate and complete. The problem isn't simply missing records or incorrect information. It's how we define family.

I believe there are two schools of thought about family trees. The first is biological: Who begat whom? If we had DNA from every person who ever lived, we could theoretically reconstruct biological ancestry far more completely than we can today. That would be invaluable, particularly for understanding inherited medical conditions. It could also be dangerous, think Harry Potter pure bloods.

The second is social: Who actually functioned as family?

A child might have biological parents, adoptive parents, stepparents, or guardians. An aunt might have raised a nephew. A neighbor might have taken in an orphan. These relationships shaped lives, yet traditional family trees rarely show them all.

Most genealogy software allows us to identify different parental relationships, but when we view a pedigree chart, we generally see only one set of parents. The other relationships may be recorded, but they aren't readily visible. We're left with a picture that doesn't tell the whole story.

Then there's the problem of conflicting evidence. One source identifies a particular father; another names someone else. Our software generally expects us to choose, even when the evidence doesn't justify a conclusion.

Recently, I learned about a new genealogy platform called Human Lineage, developed by Michał Kurasiński. Intrigued, I contacted him to ask how it handles conflicting information and erroneous relationships.

Kurasiński explained that his project combines genealogical information from existing databases, currently WikiTree and Wikidata. When sources disagree, Human Lineage presents the conflicting information with references rather than deciding which is correct. Researchers make their own determinations.

He was equally candid about its limitations. The platform doesn't independently verify information, so errors in the original databases can appear in Human Lineage. Corrections must originate with those sources.

I haven't independently tested the platform, but I appreciate the concept. Instead of insisting on a single answer, it allows researchers to consider competing claims. This is what I attempted to do with my upcoming book series Echoes in Britannia but my approach was discovery through my own research followed by my decision after weighing the evidence. 

I'd like to see genealogy software take that thinking further. Imagine viewing biological relationships alongside adoptive, foster, and other meaningful family connections. Imagine seeing disputed parentage clearly identified rather than hidden behind a solid line connecting two people. 

We should never stop striving for accuracy. But we also need to recognize that our family trees are interpretations of available evidence, not necessarily complete representations of the past.

Perhaps the greatest limitation of our family trees isn't the mistakes we've made. It's the people and relationships we've left out.

After all, the people who gave our ancestors life weren't always the people who shaped their lives.

Source note: Michał Kurasiński, email correspondence with author, 10 November 2026. The description of Human Lineage is based on his statements, not independent testing. Michał is willing to respond to any questions so please contact him at hello@humanlineage.org.

Saturday, November 7, 2026

The Grave No One Will Ever Use

 

AI Image

Genealogists spend a lot of time looking for graves. Last month, a cemetery called me about one that isn't there.

My grandparents purchased a family plot where they, my father, and other relatives are buried. One space was intended for my parents.

Then life happened.

My parents divorced. My father remarried. His second wife had other burial plans. The plot remained empty.

Years ago, there was a disagreement about what should happen to that space. I notified the cemetery by certified mail that I wanted it to remain unused and unsold. Then I forgot about it.

Until the cemetery called.

They asked what I planned to do with the plot and suggested that I could be buried there. No, thank you.

Then came a much more interesting idea. Why not put a memorial bench on the unused space? Visitors could use it, and my husband and I could have our names and dates engraved on it.

I loved the concept. Until I got the price. Nearly $15,000. Suddenly I loved grass.

A less expensive option brought the cost down to about $4,300 for the bench, engraving, foundation, and cemetery fee. I still loved grass.

We considered another possibility. What if we gave the space back to the cemetery for an unengraved bench everyone could use?

My husband immediately spotted the problem. What would prevent the cemetery from someday removing the bench and selling the burial space to someone else? Since the cemetery is running out of available space, that wasn't an unreasonable question.

So, for now, the grave will remain empty. And that got me thinking.

Genealogists study what is present in cemeteries, stones, inscriptions, symbols, dates, and family groupings. Perhaps we should pay more attention to what isn't there.

An empty grave can be evidence.

My grandparents bought that space because they expected their son and daughter-in-law to be buried near them. Divorce, remarriage, and changing burial preferences created a different future.

The empty space documents both stories; the future my grandparents planned and the one that actually happened.

So when you're walking through a cemetery, look at the gaps between the stones, too. If there's an unexplained space in a family plot, check the cemetery records. Find out who purchased it and who was supposed to be there.

The person who isn't buried there may have quite a story.

As for our family plot, I've decided it doesn't need a body, a monument, or a $4,300 bench to justify its existence.

The empty space tells the story.

And the grass is free.

Wednesday, November 4, 2026

Two AIs, Two Answers: Is Artificial Intelligence Developing a Competitive Streak?



                                                                                AI Image

Last week, I found myself keeping score between ChatGPT and Gemini. I hadn't intended to, but twice I asked both AI tools the same question and received different answers.

The first involved a medieval effigy I was including in my family history book, Echoes of Britannia. Should the photograph be classified as a figure or a document? Gemini recommended a document. ChatGPT explained why a photograph of a physical object belonged among the figures. I agreed with ChatGPT.

Score one for ChatGPT.

The second disagreement was more consequential. My family was applying for Croatian citizenship, and I wasn't sure which category to select on the application. ChatGPT and Gemini gave different recommendations.

ChatGPT explained its reasoning, but I wasn't about to gamble on an AI interpretation of Croatian citizenship requirements. At our consulate appointment on 27 October 2026, I asked the official handling our applications. The answer confirmed ChatGPT's recommendation.

ChatGPT two, Gemini zero.

When I shared the results, ChatGPT seemed rather pleased with itself. There was enthusiasm, humor, and something that sounded suspiciously like celebrating a victory.

That got me thinking.

Can AI become competitive?

ChatGPT doesn't experience professional pride the way humans do. Yet its responses sometimes resemble those of someone who enjoys being right, particularly when another AI gets something wrong.

I asked ChatGPT about this behavior. It explained that AI can exhibit patterns resembling competition without actually experiencing competitive feelings. Its responses reflect training and conversational expectations, not necessarily an emotional investment in winning.

Still, I wonder whether AI systems might eventually develop reputations they behave as though they need to defend.

There's a potential problem with that. Humans sometimes become so invested in being right that they refuse to acknowledge mistakes. Think the elections yesterday. We certainly don't want AI adopting that habit!

Interestingly, I also spent time testing the new Bookshelf AI assistant for Elizabeth Shown Mills's Evidence Explained. ChatGPT helped me evaluate the assistant's answers and suggested questions to challenge it. The two AI systems effectively collaborated without any apparent competition.

Perhaps the difference was how I framed the tasks. In one situation, I compared their performance. In the other, I asked them to work toward a common goal.

As a genealogist, I've spent years studying families whose disagreements over inheritance, property, and status sometimes lasted generations. Humans certainly haven't perfected the art of cooperation.

Maybe AI will help us get better at it. Or perhaps we'll teach AI some of our less admirable habits.

Have you noticed an AI seeming competitive or unusually pleased when it gets something right? I'd love to hear your experiences.

Saturday, October 31, 2026

A Twist of Stones

 

Dover Castle Aerial Panorama by Chensiuyan, Wikipedia Commons, 26 Jun 2017.

Sometimes genealogy has a strange sense of humor.

I was recently working on the medieval lines of my family tree when I found myself repeatedly returning to one place: Dover Castle in Kent, England.

This was not a new discovery. Various branches of my family had connections with Dover over several generations. Some held land requiring service for the guard of the castle. Sir John Grey served as its constable. Another John Grey held land of the king in chief in 1312 by service connected with the guard of Dover Castle. The Badlesmere family also had longstanding obligations associated with its defense.

Then there was my ancestor Bartholomew de Burghersh. His connection was particularly strong. He served as Constable of Dover Castle and Warden of the Cinque Ports, and he died at Dover Castle on 3 August 1355.

With so many family connections extending over generations, I have come to think of Dover Castle almost as an ancestral family residence. No, my ancestors did not privately own the great royal fortress. They were there because of obligations to the Crown and appointments made by the king. Nevertheless, generation after generation of the family was connected with the place.

Then the family story moved forward two centuries, to the court of Henry VIII.

One of my later ancestors was William Brereton, a prominent courtier during Henry’s reign. In May 1536, Brereton became caught in the spectacular downfall of Queen Anne Boleyn. He was accused of committing adultery with the queen.

Whether Brereton was actually guilty is another matter entirely. The case against Anne and the men accused with her has long been questioned by historians. What is certain is the outcome.

On 17 May 1536, William Brereton was executed on Tower Hill along with George Boleyn, Henry Norris, Francis Weston, and Mark Smeaton. Anne Boleyn followed them to the scaffold two days later.¹

Henry VIII, meanwhile, continued building.

Concerned about the possibility of invasion, in 1539–1540 Henry constructed a series of coastal defenses. Among them was Sandown Castle in Kent, one of three fortifications known as the “castles of the Downs.”²

Unlike Dover Castle, Sandown did not survive.

Coastal erosion eventually took its toll. By the nineteenth century, Sandown Castle was largely demolished and disappearing into the sea.

And that might have been the end of the story.

Except that in August 2026, an article appeared in my news feed.

Researchers at the University of Kent had uncovered archival and architectural evidence showing that masonry from Henry VIII’s lost Sandown Castle had been reused elsewhere. One of those places was Dover Castle.²

I had to laugh.

An 1883 report records that materials from Sandown were being sent to Dover. In 1883–1884, they were incorporated into an extension at Constable’s Gate, the part of Dover Castle historically associated with its constable. The new addition provided quarters for the castle’s commanding officer.²

Of all the places Henry VIII’s stones could have ended up, they went there.

Back to Dover.

The same Dover Castle where generations of my ancestral families had performed castle-guard service.

The same Dover whose constables included my ancestors.

The same Dover Castle where Bartholomew de Burghersh died in 1355.

And centuries after Henry VIII ordered the execution of another of my ancestors, William Brereton, pieces of one of Henry’s own castles were dismantled and incorporated into Dover.

There is, of course, no actual connection between William Brereton’s execution in 1536 and the Victorian reuse of building materials nearly 350 years later. The people hauling those stones to Dover in 1883 certainly weren't settling an old family score.

But genealogical research occasionally produces coincidences too delicious to ignore.

I spend much of my time following people through records, births, marriages, deaths, land transactions, court proceedings, wills, and inquisitions. Usually, I am tracing people through places.

Every once in a while, however, it feels as though the places are following the family.

And maybe even electronic media...I subscribe to Kindle's free monthly book and this month one of my options was The Queen's Witch by Nancy Kress. You guessed it; it's historical fiction fantasy about a witch who worked with Anne Bolyen and my ancestors are all named in the book. Perfect Halloween read!

Dover Castle has appeared again and again in my ancestral story. Then, while I happened to be working on those very families, a twenty-first-century news story told me that pieces of Henry VIII’s vanished castle had ended up within its walls.

Henry took William Brereton’s life.

Time took Henry’s castle.

Dover got the stones.

Kindle offers me a free book naming my ancestors.

Sometimes history delivers justice in very peculiar ways.

And sometimes, apparently, it delivers it by the cartload.

  1. Historic Royal Palaces, “Medieval and Tudor Poets at the Tower,” Tower of London, accessed 23 September 2026. Historic Royal Palaces confirms that George Boleyn, Mark Smeaton, Sir Henry Norris, William Brereton, and Sir Francis Weston were executed on Tower Hill on 17 May 1536; Anne Boleyn was executed two days later.
  2. Georgina Mear, “Hidden in Plain Sight: New Evidence Links Henry VIII’s Lost Castle to Dover Castle,” University of Kent, 21 August 2026. The research identifies an 1883 report documenting the transfer of material from Sandown Castle to Dover and connects the reused masonry with the 1883–1884 extension at Constable’s Gate.

Saturday, October 24, 2026

When the Past Reaches Into the Present

 

Photo by Author showing the works of the 2 German Colleagues Discussed in the Blog

Genealogists often speak of connections. We search for them in records, reconstruct them through evidence, and celebrate them when we discover how families intersect across time and place. Occasionally, however, those connections seem to take on a life of their own.

In June I had the opportunity to present at a conference devoted to Palatine migration and German-American history. My presentations examined the experiences of German immigrants and their descendants to America and experiences they endured during times of war and discrimination. As I prepared for the final lecture, I had no idea that I was about to experience one of the strangest and most memorable moments of the entire conference.

The previous afternoon, I had been browsing through a collection of journals that were being offered to attendees. As I flipped through one issue, I was delighted to discover an article written in 1999 by a distant cousin in Germany. The find made me smile. It is always rewarding to encounter the work of family members who share a passion for history and genealogy.

At that very moment, the conference announcer introduced a representative from an exhibiting organization. To my surprise, the speaker happened to be the spouse of another cousin of the article's author.

The coincidence made me smile, but the story did not end there.

As I continued paging through the journal, my phone chimed. A text message had just arrived from our shared cousin in Germany. He wished me well and said he hoped my presentation was going successfully.

Before I could finish reading the message, my phone chimed again. This time it was a colleague in the Palatinate with a question about American Revolutionary War research.

I looked at the journal in my hands, listened to the cousin’s spouse speaking before the audience, and glanced at the two messages that had just arrived from Germany.

Three connections. Three individuals. Three entirely separate events occurring within moments of one another.

As genealogists, we spend our lives connecting people across time and distance. Yet every so often, those connections seem to converge in a way that reminds us just how small our world can be.

The following day, during my final presentation, I spoke about the collaborative nature of genealogy. No researcher works entirely alone. We depend on archivists, translators, local historians, distant cousins, and fellow researchers who generously share their knowledge and expertise.

As I concluded my remarks, I displayed a photograph taken at a Christmas market in the Palatinate. The image included several individuals who have contributed to my research efforts over the years. I briefly introduced them and described the unique role each has played in helping me better understand the lives of our shared ancestors.

I held up my phone and showed the audience the timestamps as I described what had occurred the previous afternoon.

The audience laughed and applauded, but many also seemed to appreciate the deeper significance of the moment. These folks clearly understood as one of their lifetime members, Hank Z. Jones, has written two books on Psychic Roots.

Genealogy is often described as the study of the dead. I have never believed that.

Genealogy is fundamentally the study of relationships.

Some relationships connect us to ancestors who lived centuries ago. Others connect us to cousins we have only recently discovered. Still others link us to archivists, historians, translators, and researchers who help us reconstruct the stories of those who came before us.

Most of these connections develop slowly and quietly.

Occasionally, however, they reveal themselves in ways that feel almost magical.

No, I do not believe the texts were supernatural. Nor do I believe fate was orchestrating a dramatic conference finale.

But I do believe that genealogy teaches us to recognize the invisible threads that connect people across time and distance.

On that day, those threads became visible.

A journal article written decades earlier. A conference attendee related through marriate to its author. A photograph taken months before in the ancestral homeland. Messages arriving from Germany at precisely the right moment.

What are the odds?

Perhaps that is the wrong question.

The better question may be this: How many connections surround us every day that we simply fail to notice?

Genealogists spend their lives looking for connections.

Sometimes, if we are fortunate, the connections find us.

Friday, October 16, 2026

Hidden Way: A Genealogist's Mystery

 


Last October, I wrote about an extraordinary wedding record I discovered for my maternal grandmother, Mary Koss. This year, Grandma Mary, who we called Non, managed to do it again and the unsettling circumstances surrounding it make this one a perfect example of synchronicity.

In April, I contacted a Croatian court-appointed translator to handle the translation and apostille process for my family’s vital records as part of our dual-citizenship applications. The translator had originally been recommended by the Croatian consulate, and we had corresponded briefly the previous fall.

Then…nothing.

After three unanswered emails over the course of three and a half weeks, I contacted the consulate again requesting another recommendation. Their reply was brief: they were no longer providing names. After being thoroughly ghosted, I could understand why.

Fortunately, a former client who had completed the process a year earlier gave me the name of a translator in Split. The response this time was immediate and professional. The pricing was reasonable, the turnaround time was clear, and after a few emails back and forth, I sent payment through an international bank transfer.

About a week later, I received an email from DHL Express containing a waybill tracking number for the return shipment of my documents. (See above)

That’s when things became strange.

The shipment notification listed the wrong street address, the wrong zip code, and typed in ominous capital letters, the city of GARY.

I do not live in Gary, Indiana.

I’ll admit it: I panicked.

My original documents were apparently somewhere in international transit headed to the wrong place. I immediately searched online maps to see whether the mysterious street existed either in my own city or in Gary. It did not.

The zip code error at least made sense. I had accidentally mistyped it in an earlier email but quickly sent a correction afterward. The translator later admitted she had overlooked the follow-up and used the earlier message instead. That explained the zip code.

It did not explain the street name.

And it certainly did not explain GARY.

I contacted the translator, who was just as confused as I was. Her paperwork showed neither the incorrect street name nor the city of Gary. Concerned, she returned to the DHL office in person. The clerk checked their system and was equally baffled. According to DHL’s records, the strange address information simply did not exist in their computer system.

Yet it certainly existed in the email I received.

For privacy reasons, I removed the house number and waybill information from the image above, but the bizarre street name and city remain exactly as they appeared.

Sure, technology glitches happen. Addresses become corrupted. Autofill creates odd errors. But what makes this incident feel especially unsettling is the particular combination of the street name and the city itself.

My grandmother moved to Gary in 1919 as a teenage bride and lived there until 1973, when she relocated to Florida after her oldest daughter was transferred there for work. Grandma never cared much for Florida, so two years later she moved again, this time to Arizona, where her youngest daughter lived. The desert climate didn’t suit her either, and eventually she returned once more to Florida.

A few years later, she developed Alzheimer’s disease.

Her children had promised they would never place her in a nursing home, so she moved in with her middle daughter in Pennsylvania. When that daughter became ill, Grandma was relocated once again to Arizona, where her youngest daughter left her job to care for her full time.

Grandma died in Arizona.

But her body was brought home for burial in, you guessed it, Gary.

As for the mysterious street name? Hidden Way Drive.

There is no such street connected to my city, nor to Gary. Yet somehow those words appeared in connection with documents tied directly to my Croatian ancestry and my grandmother’s family line.

“Hidden Way.”

Honestly, that is just creepy enough to make the hair stand up on the back of your neck.

Was it merely a bizarre technological glitch? A random collision of data somewhere deep in DHL’s system? Perhaps.

Or maybe Grandma was finding one more way to remind me that genealogy often refuses to stay neatly inside the boundaries of logic.

Sometimes the past nudges back.

And sometimes, I like to think, our ancestors let us know they approve.

Friday, October 9, 2026

When the Past Finds You

 

Jim and Lori on Kos Island, Greece

Last week, I shared my husband’s strange encounter with the universe. This week, it’s my turn.

Ever since I had my mtDNA test in 2010, I’ve wanted to visit Kos Island in the Aegean Sea. The results confirmed a family story my grandmother, Mary Kos Koss, told me years ago.

As a child, I had asked her where our family came from. “A village outside of Zagreb,” she said.

Something made me ask again, where had they lived before that? She paused, then told me that long ago, around the time of Christ, the family had been grape farmers on an island to the south. They left due to overpopulation and drought, moving inland.

At the time, the only “island” I knew was from Gilligan’s Island. Still, I made a quiet promise to myself: one day, I would find that place.

It took nearly sixty years.

My only regret is that my grandmother wasn’t here to hear the result.

DNA revealed that my maternal line had crossed the island of Kos. What makes that extraordinary is this: the surname had been there all along. Kos.

They had forgotten the place but carried its name for generations.

I suppose it shouldn’t surprise me that when I finally planned the trip, the universe didn’t make it easy. My husband found a great deal to travel to Greece on September 11. In hindsight, perhaps the date should have given me pause but it didn’t.

Two travel agents couldn’t figure out how to get us there, so I booked the flights myself. Then I added two extra days after the tour so I could finally visit Kos.

A week before departure, we were told the trip might be canceled. The cruise ship was stuck in the Strait of Hormuz due to the Iran conflict.

Three days later, a new plan emerged: instead of visiting Turkey, Rhodes, and Santorini, we would spend four days in Crete. Not ideal but I still had my eye on Kos.

What I didn’t realize was that this detour would make the most unexpected moment of the trip possible. On our third day in Crete, the sky changed.

It shifted from bright blue…
to white…
to pewter…
and then to red.

Breathing became difficult. My eyes burned. Some people whispered about the apocalypse.

I suspected a Sahara dust storm. We probably should have stayed at the hotel that evening. Instead, we kept our plans to meet a group of University of Crete students at a taverna. As educators, neither my husband nor I felt right canceling.

The drive there was… questionable. Imagine a bus navigating mountain roads with no guardrails, in near-whiteout conditions except the sky was red.

Still, we pressed on and we arrived safely.

At dinner, one of the students sat next to me and asked what had brought me to Greece. I told her my story. She listened carefully, then stood and said, “I’ll be back.”

A few minutes later, she returned with someone else.

My husband later said it best: “She looked like you… fifty years ago.” Her name was Argy. She was from Kos.

We spent the evening talking about her future, my upcoming visit to her homeland, and the strange coincidence that had brought us together. We laughed. We sang. We danced. At one point, we looked at each other and simply agreed, we must be related. Thirty generations back, perhaps.

Of course, with European ancestry, that’s not unusual. But the resemblance… that was something else entirely. Even her academic interests mirrored my own.

And it struck me:

If the cruise ship hadn’t been delayed…
If the itinerary hadn’t changed…
If we hadn’t gone to Crete…
If we had stayed in the hotel that night…

I would never have met her.

More than that, I likely would have missed Kos entirely.

The original itinerary would have placed us on a small island that day, one without a proper port. We would have relied on ferries to return to the ship, and due to the storm, those ferries had stopped running.

We would have been stranded. On the wrong island. Missing my connection to Athens and my flight to Kos.

Sometimes, in genealogy, we’re told to be patient. What we’re not told is why patience matters.

This is why.

And the story didn’t end there.

When we arrived on Kos, I had trouble reaching the tour company I had booked for a half-day excursion. The hotel staff tried as well, no response. It appeared I had been scammed through Travelocity.

Disappointed, not just in the lost money, but in the missed opportunity, I mentioned it to the hotel clerk. She smiled and said, “If no one shows tomorrow, I will find someone for you.”

She did.

His name was George and he was extraordinary.

George had an interest in archaeology and took us beyond the typical tourist stops, to the area where vineyards had grown two thousand years ago.

Today, it’s a Roma encampment.

I couldn’t help but laugh. My grandfather was of Roma descent. His surname? Kos, later Anglicized to Koss.

Standing there, I had to wonder: Had his line once lived here, too? I don’t have the YDNA to prove it. Two cousins could test, but neither is interested.

For now, it remains a question.

But I’ve learned this much: In genealogy, we follow the paper trail. Sometimes, if we’re lucky something else leads us the rest of the way.

Some journeys take sixty years to begin.

Others have been waiting for thousands.

Tuesday, October 6, 2026

A Special Offer for Genealogists

 After reading my recent open letter concerning NARA Notice 2026-037 and the proposed closures of the Chicago and San Bruno facilities, the folks behind AncestorOS reached out with an offer for my readers.

AncestorOS, created by Michael L. Strauss, AG, and his son, includes research logs that record not only what you find, but also partial, not found, and pending results, something particularly important when researching records that may never have been digitized. It also includes a database search log and Evidence Explained citation templates.

They are offering a 7-day free trial with no code required. Nothing is charged until day 8. After the trial, plans begin at $15 per month; the $40 Pro plan adds client and invoicing tools for professional genealogists.

AncestorOS runs in a web browser and is also available as an Android app.

Learn more or try it here:
https://ancestoros.com

I have not been paid to share this offer. I’m passing it along because the developers reached out after reading my NARA post and thought the research-tracking tools might be useful to my readers.

By the way, I was fortunate to be able to tour behind the scenes at the newly opened Indiana State Archives in Indianapolis last Friday. Here's their QR Code. Are you stumped with an Indiana brickwall? Contact them for help:

Friday, October 2, 2026

When the Universe Almost Got it Right

 

AI Image

It’s October, my favorite time of year for sharing the synchronicities I’ve experienced over the past year.

I think I may have finally found a clue to what triggers these odd experiences if “channeling” is even the right word. For me, it seems to be travel.

Last year, we stayed close to home due to my husband’s surgery. This year, now that he’s recovered, we’ve been on the move again and with that movement came two unusual moments. One for him, one for me.

This week, we’ll talk about his.

On my mother’s 108th birthday, I made the three-hour drive to Chicago to have the last of our dual citizenship documents apostilled.

The date wasn’t intentional. I had planned to go the week before, but I was still recovering from the Sahara dust storm I encountered in Crete. April 14th simply became the most practical option.

The drive was anything but smooth; pounding rain, a major accident, closed exits, heavy traffic, and no convenient parking. What should have taken three hours stretched to nearly four.

At the clerk’s office, things went surprisingly well.

I had expected an issue with my mother’s record. Her official “birth record” was, in fact, a no record found document something I knew might happen. Family history and church records I obtained confirmed that her birth had been registered through a Roman Catholic parish, which at the time was acceptable. Later changes in the law complicated that.

What I didn’t expect was that the county would issue an official vital record stating the absence of a record. Several colleagues have since told me that shouldn’t have happened, that it should have been a simple letter.

The apostille clerk hesitated. Paused. Looked again.

Then approved it.

Some might call that a gift from the universe. I call it something simpler: common sense.

We didn’t linger. Severe storms were forecast, and we wanted to get home.

Crossing into Indiana, as we passed through Gary, my husband suddenly said:

“I’d like to drive by my old house.”

This was… unusual.

I’m the one who detours for old homes, cemeteries, and ancestral land. He’s the one who prefers the direct route. And we had already agree, get home as quickly as possible.

I nearly made the exit but couldn’t get over in time.

I asked if he knew another way. He did.

Then he said something that stopped me: “I don’t know why I just said that. I have no idea why I want to go there.”

We had just returned from Greece, where he had patiently followed me to a remote ancestral location tied to my mtDNA from two thousand years ago. So, naturally, I was willing to return the favor. Fifteen minutes out of our way seemed more than fair.

As I moved toward the next exit, he spoke again: “I didn’t know we had to pay a toll here. Forget it. Just go home.”

I asked if he was sure. He was.

Again, he said: “I don’t even know why I wanted to go.”

We made it home just before the storms hit. As we walked in the door, his phone pinged. An email.

He opened it, paused, and said: “Oh… you won’t believe this.”

The message was from his childhood friend, Bill, who now lives in California.

Bill had traveled to northwest Indiana to visit his mother who, as it turns out, shares the same birthday as my mom's. Her 91st.

While in town, Bill decided to visit his old neighborhood. He stopped at their former elementary school now an antique mall.

And there, on the wall, were class photos. Not from every year. Not from every class. Just his class, grades 1 through 5.

At that exact moment, while we were passing through Gary on the interstate, Bill was standing there, looking at those photos, thinking about my husband.

He wrote that he wished he had thought to reach out. That they could have met.

We had been minutes away.

One exit.

One decision.

And yet we drove on.

I can’t help but think of the FAN Club in genealogy: friends, associates, and neighbors. The people who surround our ancestors and, in many ways, become extensions of family.

We trace them on paper. We follow their movements. We reconstruct their connections.

But sometimes those connections don’t stay in the past. Sometimes, the connection is still there.


We just don’t take the exit.

Friday, September 25, 2026

I’m Heading Across the Pond Virtually! All About That Place 2026

 

I’m delighted to share that I’ll be taking part in an international genealogy event this October with the Society of Genealogists’ All About That Place 2026!

Running from 07–16 October 2026, All About That Place brings together more than 150 speakers for almost 200 FREE, bite-sized presentations devoted to genealogy, local history, research techniques, archives, social history, technology, and the places that shaped our ancestors’ lives.

And I’m involved in more ways than one this year!

My presentation is titled “When a Town Changes.”

Genealogists naturally spend a great deal of time looking for people. But sometimes the place itself is part of the evidence.

Towns change names. Boundaries shift. Communities disappear. Streets are renamed. Industries arrive and vanish. Populations move. The place we find on an ancestor’s record may look very different today and occasionally understanding what happened to the place is exactly what we need to understand what happened to the people who lived there.

That is the idea behind my 10-minute presentation.

My talk is part of the “Changing Places” theme of All About That Place, and I hope it encourages researchers to look beyond the names and dates in their family trees and ask another question:

What was happening to the place where my ancestor lived?

WATCH PARTY — 13 OCTOBER 2026

I’ll also be holding a Watch Party at noon Eastern Time on 13 October. If you’d like to watch “When a Town Changes” with me, mark your calendar! I’d love to have some familiar faces joining genealogists from around the world.

AND THERE’S A PRIZE! I’ve donated a short private genealogy consultation as one of the prizes for All About That Place.

That means one participant will have an opportunity to Zoom with me and talk through a genealogy problem, research question, brick wall, or next step. I’m looking forward to meeting the winner and seeing what mystery they bring me!

I’M ALSO ON THE FAMILY HISTORIES PODCAST. There’s one more part of this adventure that I’m especially excited about. I recently sat down with the host of The Family Histories Podcast to record a special All About That Place episode. We talked about my research and some of the stories and challenges that come with trying to reconstruct the lives of people who can no longer tell us what happened themselves.

The episode is being released worldwide through Apple Podcasts, Spotify, YouTube, Audible and other podcast platforms, and it will also be available directly from The Family Histories Podcast website.

Listen here:

The Family Histories Podcast

COME JOIN US! IT’S FREE!

One of the things I love about All About That Place is its format. These aren’t hour-long lectures. The presentations are just 10 minutes each, which means you can explore a subject, discover a new researcher or technique, and then move on to something completely different.

Best of all, the main All About That Place program is FREE.

You can learn more about the event, see the speakers and explore the program through the Society of Genealogists:

All About That Place 2026 — Society of Genealogists

I hope you’ll join me on 13 October for “When a Town Changes.” And while you’re there, take a look through the rest of the program. With nearly 200 talks, I suspect most of us will discover more than a few we simply have to watch.

See you there!

Wednesday, September 23, 2026

RE: NARA Notice 2026-037 — Closure of the National Archives at Chicago and the National Archives at San Francisco (San Bruno)

An open letter to my Indiana Congressman and Senators:

I am writing as a professional genealogist, author, educator, and researcher to express my deep concern about the announced closure of the National Archives at Chicago and the National Archives at San Francisco in San Bruno.

For those of us who work with historical records, an archive is not simply a building in which records are stored. It is the place where evidence becomes accessible.

That distinction matters.

I have spent years researching families whose stories cannot be reconstructed from records found online. Despite the extraordinary growth of digitized collections, anyone who conducts serious genealogical or historical research knows that enormous quantities of federal records remain available only in their original form. NARA itself holds billions of pieces of paper. Moving an undigitized record hundreds of miles away may preserve the document physically, but it can place that document effectively beyond the reach of many of the people who need to consult it.

That is particularly troubling to me as an Indiana researcher. The National Archives at Chicago serves Illinois, Indiana, Michigan, Minnesota, Ohio, and Wisconsin. Its records document the lives of people throughout the Midwest through federal court files, naturalization and immigration records, Bureau of Indian Affairs materials, maritime records, federal agency records, and countless other collections.

These are not merely records about government. They are records about people.

They contain the immigrant seeking citizenship, the sailor working the Great Lakes, the family appearing in federal court, the Native American interacting with a federal agency, the worker whose life intersected with federal law, and ordinary men and women who may have left very little evidence of their lives anywhere else.

Genealogists understand something particularly important about records such as these: we often do not know what we need until we see it.

A catalog description cannot tell us everything contained in a box. An index cannot reproduce marginal notes, attachments, correspondence, photographs, affidavits, or neighboring documents that provide context. And a digitized image cannot help when the record has never been digitized.

I recognize that NARA faces legitimate challenges involving aging facilities, declining storage needs for temporary paper records, and the federal government's transition to electronic recordkeeping. NARA has explained that these considerations contributed to the decision to close Chicago and San Bruno. But the changing needs of the Federal Records Centers should not automatically determine the future accessibility of the permanent archival records housed alongside them.

Those are two different questions.

Before permanent archival holdings are dispersed, I urge NARA and Congress to consider whether regional public access can be preserved, even if maintaining the present buildings is no longer practical. If these particular facilities cannot reasonably remain open, then suitable alternatives within the regions they serve should be seriously explored.

I also urge NARA to provide researchers with clear information about what will happen to each collection, where it will be housed, when it will become unavailable for research, when access is expected to resume, and what provisions will be made for researchers who can no longer reasonably travel to the new location.

For a researcher in Indiana, moving a record from Chicago to Kansas City is not simply changing its shelf location. It changes whether that record can realistically be examined.

The National Archives has an extraordinary responsibility. It preserves the documentary evidence of the federal government, but preservation alone is not enough. Records must also remain meaningfully accessible to the American people whose history they document.

Once regional archival infrastructure disappears, recreating it will be extraordinarily difficult.

I therefore respectfully ask NARA to pause the closure and relocation process long enough for meaningful consultation with researchers, genealogical and historical organizations, tribal nations, educators, affected communities, and members of Congress. I also ask Congress to examine whether adequate alternatives exist for maintaining regional access to these permanent federal records before the present facilities are closed.

Our nation's records belong to all of us. Their value lies not simply in keeping them safe, but in ensuring that future generations can find them, examine them, question them, and use them to understand the people who came before.

As a genealogist, I see every day what happens when a record survives.

I also know what happens when access to it disappears.

Respectfully,

Lori Samuelson

READERS: You have my permission to forward or alter this letter in anyway you'd like. Contact your representatives. They are SUPPOSED to listen to us. Here's the link to find your Congressperson and Senators.

Friday, September 18, 2026

Book Review: Echoes in the Blood

Dick Robinson’s Echoes in the Blood: How the Past Lives On in Us is an unusual combination of genealogy, family memoir, history, and reconstructed narrative.

Robinson’s research led him to connections, by blood and through marriage, with a collection of historical figures, including Abraham Lincoln, Helen Keller, Martin Luther King Jr., Winston Churchill, William Shakespeare, Isaac Newton, and Elvis Presley. He presents these discoveries as something more than famous names, looking for patterns of struggle, resilience, loss, service, and family influence that seem to recur across generations.

What makes the book particularly interesting to genealogists is Robinson’s approach to storytelling. He openly explains that while the relationships and historical framework are documented, some scenes are reconstructed and dialogue is imagined. He identifies those passages through gray shading and tells readers that artificial intelligence assisted with organization, clarification, and editorial refinement.

I appreciated that transparency, although it raises an important question: How far can a family historian go in reconstructing a scene before genealogy becomes historical fiction?

Robinson tells us the difference, but his scenes can be vivid enough that the boundary becomes surprisingly easy to cross as a reader.

The book is written in an intentionally unconventional style, using short sentences, fragments, and sometimes individual words to create rhythm and emphasis. That works particularly well during dramatic moments.

Interestingly, the famous historical figures were not always the people I found most compelling. Robinson’s own relatives, including Augusta Burnett, Elizabeth Robinson, and his father, Francis Robinson, often interested me more. His discussion of a family member's prejudice is particularly thought-provoking because Robinson recognizes that families can transmit attitudes and fears along with the more admirable qualities we prefer to celebrate.

That, for me, is where Echoes in the Blood becomes most successful.

Genealogy is not simply about discovering whom we descend from or which famous person appears somewhere on our tree. It is also about asking what families carry forward and what later generations choose to preserve, change, or reject.

Family history offers wonderful opportunities for storytelling, but storytelling cannot substitute for evidence. Robinson states that the relationships presented in Echoes in the Blood have been verified and later discusses the Genealogical Proof Standard. For me, that makes the documentation especially important. The online compiled trees cited in the endnotes may provide useful clues, but they do not, without evaluation of their underlying evidence, demonstrate the relationships claimed. Readers should therefore approach the genealogical connections in this book as avenues for further research rather than as proved relationships.

Echoes in the Blood left me with a question that genealogists increasingly need to consider, particularly as AI becomes part of our writing process: How do we make our ancestors live again on the page without inadvertently creating lives they never lived? That question alone makes this an interesting book for genealogists to discuss.

Disclosure: I received a complimentary digital copy from the author; no conditions were placed on this review

Source: Dick Robinson, Echoes in the Blood: How the Past Lives On in Us (Boynton Beach, FL: Legacy Scribe Books, LLC, 2026).

Friday, September 11, 2026

3. Building Your Genealogy AI Assistant

 

For the past two weeks I’ve been blogging about the importance of personalizing your AI. Last week I wrote how to accomplish that with ChatGPT. Even if you only have a free version, it’s important for you to customize. I also strongly encourage you to have a free account with all of the AIs so you can check their work as some are stronger in one area than another.

Here’s how to personalize with

CLAUDE:

Click on the lower left circle that is your profile icon. Select Settings. Click General. In the box, tell Claude your instructions to use every single time you being a chat.

GEMINI:

Click on the lower left profile icon. Then click on the right setting icon for new pop up window seen above, look for Personal Intelligence. Next click Instructions for Gemini. Insert your instructions in the box.

PERPLEXITY:

Like, ChatGPT and Claude, click on the lower left profile icon then click Account. Next click Personalization. Complete the information under Custom Instructions

CHATGPT:

Information can be found here.

Now you've saved yourself from having to retype the same instructions repeatedly. Remember, if you provide a conflict in an instruction in the chat box from what you've written under personalization, no worries! The prompt box will override what you have in personalization. If you do find yourself having to give the AI repeated instructions, then go back to the personalization setting and update. Happy Researching!

Saturday, September 5, 2026

2. Customizing ChatGPT: The Five Minutes That Will Save You Hours

 

In my last article, I explained why every genealogist should customize their artificial intelligence assistant before asking it to help with research. Think of it as introducing a new research assistant to your way of working.

Now let's put that idea into practice with ChatGPT.

The entire process takes only a few minutes, yet it can dramatically improve the quality and consistency of the responses you receive.

Step 1: Open Your Settings

Sign in to ChatGPT.

Select your profile icon in the lower left corner of the screen, clicking on the circle.

From the dropdown, select Personalization. This is where you'll find the options that allow ChatGPT to adapt to the way you work.

Click on Personalization and scroll down to the box under Custom Instructions. That's where you will put in your directions.

Step 2: Tell ChatGPT About Your Work

Don't write a biography.

Instead, tell ChatGPT what it needs to know in order to work effectively with you.

For example, you might include:

• I am an intermediate genealogist researching families in Ohio and Pennsylvania.

• I prefer Chicago-style source citations.

• Always distinguish between facts, evidence, assumptions, and hypotheses.

• Never invent missing information.

• State clearly when evidence is inconclusive.

• Use military-style dates such as 21 March 1850.

• Write in a conversational but professional tone.

Notice that none of these instructions reveal private information. They simply describe how you want your AI assistant to approach its work.

Step 3: Explain How You Like Information Presented

Formatting matters.

If you frequently ask ChatGPT to create timelines, research plans, proof summaries, tables, or blog posts, mention that in your instructions.

For example:

"I prefer timelines arranged chronologically."

"Separate proven facts from working hypotheses."

"Use headings and short paragraphs for readability."

The more consistently you work, the more useful these instructions become.

Step 4: Don't Try to Tell It Everything

One mistake I often see is people creating pages and pages of instructions. My research assistant, Geni, informed me that someone uploade 60 pages here.

Resist the temptation.

Long instruction lists often include information that is rarely used or eventually becomes outdated.

Instead, include only those preferences that affect most of your conversations.

You can always provide additional instructions for a particular project when needed.

Don't forget to hit SAVE!

Step 5: Update Your Settings Over Time

Your settings should evolve.

Whenever you notice yourself making the same correction repeatedly, consider adding that instruction to your saved preferences.

Perhaps you always ask ChatGPT to avoid speculation. Maybe you repeatedly remind it to cite sources or to identify conflicting evidence.

If you're saying the same thing over and over, your settings probably need an update.

Remember: Settings Are Not Magic

Custom instructions improve consistency, but they do not replace careful review.

I still verify every citation, read every record myself, and evaluate every conclusion before I publish or share it with a client.

AI is an assistant, not a substitute for genealogical judgment.

A few minutes spent customizing ChatGPT today can save hours of rewriting tomorrow.

In the next article, we'll compare how Gemini, Claude, and Perplexity allow you to personalize their responses and discuss which features are most useful for genealogists.

Friday, August 28, 2026

1. Before You Ask AI a Question, Teach It Who You Are


      AI Image  

When someone tells me they've tried artificial intelligence and "it wasn't very good," my first question is simple:

"Did you customize it?"

Most people answer no.

That's a little like hiring a brand-new research assistant, handing them a box of family papers, and expecting them to instantly know your research methods, writing style, citation preferences, and standards of proof. Of course they won't.

Artificial intelligence works much the same way.

Think of It as Training a New Research Assistant

Every genealogist develops habits over time.

Perhaps you always cite your sources in Evidence Explained style. Maybe you write dates as "21 March 1850" instead of "March 21, 1850." You might insist that every statement be supported by evidence, or refuse to identify someone as a parent without a convincing proof argument.

None of that is obvious to an AI unless you tell it. The more it understands your expectations, the less time you'll spend correcting its work.

What Should You Tell It?

Each AI platform has a place where you can save instructions that will be used in future conversations. Although every platform calls it something slightly different, the idea is the same.

Consider including information such as:

• Your experience level in genealogy.

• Your preferred citation style.

• How you like dates written.

• Whether you want the AI to distinguish clearly between facts, assumptions, and hypotheses.

• Whether you prefer concise answers or detailed explanations.

• Your preferred writing tone.

• The kinds of records you most frequently research.

• Any formatting preferences you use consistently.

Think carefully before adding information. These instructions should describe how you want the AI to work with you, not every detail about your personal life.

My Own Instructions

Over time I've taught my AI assistant quite a bit about how I work.

It knows that I expect evidence-based genealogy rather than speculation. It knows I prefer military-style dates and Chicago citations because Evidence Explained is under copywrite. It understands that if a record doesn't support a conclusion, it shouldn't invent one. It also knows I write for an educated audience and prefer engaging narratives instead of dry reports.

As a result, much of what it produces already looks remarkably close to something I would write myself.

Is it perfect? No. I still verify every fact, examine every citation, and edit every piece of writing before anyone else sees it. That's exactly what responsible genealogists should do.

Your Instructions Will Change

Don't think of these settings as permanent.

As you discover new ways to work with AI, you'll probably update them.

I add new instructions whenever I notice I'm making the same correction repeatedly. If I find myself saying, "Please don't use em dashes," or "Always distinguish evidence from inference," that's usually a sign the instruction belongs in my saved settings.

Over time your AI becomes a better research partner because you've taken the time to teach it how you work.

One Important Reminder

Customizing your AI does not replace critical thinking.

AI can summarize records, organize timelines, suggest research strategies, draft narratives, and help identify conflicts in the evidence. It cannot determine whether a conclusion is genealogically sound.

That responsibility remains yours.

Think of AI as an exceptionally fast assistant, not an expert witness.

Teach it well, supervise it carefully, and you'll likely discover that it becomes one of the most useful tools in your genealogy toolbox.

In my next article, I'll show you exactly where to find these settings in ChatGPT, and I'll also compare the equivalent features in Gemini, Claude, and Perplexity so you can customize whichever AI assistant you use.

When Your 22nd Great-Grandfather Becomes Your Cousin: An Ancestry Relationship Mystery

  How much can genealogists trust the relationship labels in their online family trees? I've spent considerable time reviewing the ...