Ep. 25:: Tracking Australian Ancestors Using AI | International Genealogy Research

Discover how to find Australian convict ancestors and track family across international borders using free AI tools. Perfect for genealogists researching British, Irish, Scottish, or Australian family history.

What You'll Learn: Track transported convicts from Britain to Australia (1788-1868) using AI-powered research techniques that work for ANY international genealogy challenge, German ancestors to Brazil, Irish to Canada, Italian to Argentina, or Scots to New Zealand.

Featured Case Study: A London weaver convicted of theft in 1832 disappears from British records. Using three free AI tools, we uncover his complete Australian life: ticket of leave, marriage to a free settler's daughter, four children, land grant, and burial in Bathurst. Plus, how DNA testing revealed American cousins who never knew their Australian family existed.

Free AI Tools Demonstrated:

• Perplexity - Research foreign record systems with citations (Australian convict terminology, record types, repositories)
• Claude - Analyze multiple historical documents, create timelines, identify discrepancies across convict indents, tickets of leave, certificates of freedom
• Gemini AI Studio - Transcribe handwritten 1800s documents with 98% accuracy (NOT the Gemini app, critical difference explained)
• NotebookLM - Create shareable infographics, audio overviews, and visual family stories

Why This Matters for American Genealogists: 162,000 convicts transported to Australia left families in Britain who became American families. If your British/Irish ancestor "vanished" 1788-1868, they may be in Australian convict databases. Learn the exact records to search: convict indents, tickets of leave, certificates of freedom, New South Wales marriage/death indexes, land grants, and cemetery records.

Key Australian Resources: State Library of New South Wales, National Archives of Australia, FamilySearch (free), TROVE newspaper archive, Colonial Secretary correspondence, church records, and DNA matching strategies for Australian cousins.

Cross-Border Research Methodology: Step-by-step framework for researching ancestors who crossed international borders. Understand destination country record systems, bridge terminology gaps (British "transport" vs Australian "ticket of leave"), locate digitized records, verify with primary sources, and maintain genealogical proof standards throughout AI-assisted research.

Perfect For: Genealogists researching British Empire migrations, convict ancestry, DNA mystery matches, international record searches, or anyone with ancestors who crossed borders 1700s-1900s.

Free Resources Mentioned: FamilySearch, Ancestry (discussed), State Library of NSW digitized collections, National Archives of Australia searchable databases, TROVE, Google AI Studio (free), Claude.ai (free tier), Perplexity (free tier), NotebookLM (free).

Keywords: Australian convict records, genealogy AI, British family history, international genealogy, DNA matches Australia, free genealogy tools, transportation records, New South Wales records, FamilySearch, Ancestry research, convict ancestors, AI genealogy research, family history podcast

Connect with Ancestors and Algorithms:

📧 Email: ancestorsandai@gmail.com
🌐 Website: https://ancestorsandai.com/
📘 Facebook Group: Ancestors and Algorithms: AI for Genealogy - www.facebook.com/groups/ancestorsandalgorithms/

Golden Rule Reminder: AI is your research assistant, not your researcher.

Join our Facebook group to share your AI genealogy breakthroughs, ask questions, and connect with fellow family historians who are embracing the future of genealogy research!

New episodes every Tuesday. Subscribe so you never miss the latest AI tools and techniques for family history research.

Transcript

So, picture this. You're scrolling through a family tree someone shared and you see a note that says, Thomas Bennett transported to Australia for theft, 1832. And that's it. That's all the information. No return date, no death date, no mention of what happened after he arrived in New South Wales. Now, if you're like most genealogists, you might think, well, that's the end of that line. He was a convict. I'll never find out what happened to him. But here's the thing. And this is what blew my mind when I started digging into Australian convict records. The British government was obsessed with paperwork. I mean, obsessed. They documented everything about these transported convicts, where they came from, what they stole, where they were assigned to work, when they got their freedom, who they married, where they moved. So the question isn't, can we find Thomas Bennett? The question is, how do we find him among the millions of pages of Australian records? And that's where AI becomes not just helpful, but absolutely game changing. Today, I'm going to show you exactly how I use three completely free AI tools to track down what happened to a transported convict and discovered a family line that nobody knew existed. And here's the kicker. And here's the kicker. Even if you don't have Australian ancestors, the techniques I'm teaching you today work for any ancestor who disappeared into another country. I'm talking about Irish immigrants who went to Brazil, I'm talking about Irish immigrants to America, Germans who went to Brazil, Scots who ended up in Canada. The methodology is the same. So stick with me because by the end of this episode, you're going to know how to tackle these international research challenges with AI as your research assistant. Welcome to Ancestors and Algorithms, where family history meets artificial intelligence. G'day, mate. I'm your host, Brian. And today we're crossing the ocean literally to solve an Australian transportation mystery. Now, I know what some of you might be thinking. I don't have Australian ancestors. Why should I care about this episode? Here's why. According to recent genealogy statistics, about 20% of Australians are descended from convicts. But here's what most people don't realize. Many of those convicts had families back in Britain who never knew what happened to them. And those British families? They became American families, Canadian families, New Zealand families. So when you're researching your English, Irish, or Scottish ancestors and someone just vanishes from the records around 1788 to 1868, there's a decent chance they were transported. And today's techniques will help you find them. Plus, the AI research workflow I'm teaching you today works for any international genealogy challenge. It's the same process whether you're tracking someone from Italy to Argentina, Poland to Australia, or Ireland to America. So let's get started with a real mystery, or at least one based on very real patterns I've seen in Australian convict research. Let's dive in. 

About three months ago, I was working with a listener, we'll call her Sarah, who contacted me through our Facebook group. She had been researching her English family line and hit a brick wall with her fourth great-grandfather, Thomas Bennett. Here's what she knew. Thomas Bennett, born 1808 in London, married Mary Clark in 1829, had one daughter, Elizabeth, born 1830, and then nothing. He completely disappears from English records after 1831. No death record. No later census entries. No burial. Just gone. Sarah had spent two years searching every database she could think of. She checked workhouse records, thinking maybe he'd fallen on hard times. She'd searched for his name in hospital records, prisons, everything. Nothing. Then, one day, she decided to search the Australian convict records on a whim. And there he was. Thomas Bennett, age 24, convicted at the Old Bailey in January 1832 for stealing cloths, sentenced to seven years' transportation, departed England aboard the convict ship, Florencia, in March 1832, arrived in Sydney in July 1832. Now, most people would stop there and think, well, at least I know what happened to him. But Sarah wanted more. She wanted to know, did he survive his sentence? Did he ever return to England to see his wife and daughter? Or did he stay in Australia and build a new life? And here's where it gets really interesting. Because when Sarah started looking for what happened to Thomas after his arrival in Australia, she found nothing. The Australian records showed he arrived. They showed he was assigned to a work gang, building roads in the Blue Mountains. And then the trail went cold. She emailed me asking, Can AI help me figure out what happened to Thomas after 1832? I feel like I'm so close. But I'm stuck again. And I told her, Sarah, not only can AI help you, I think we can solve this completely. Let me show you how. 

Now, before we dive into the solution... Let me explain why this research pattern matters, even if you don't think you have Australian ancestors. First, between 1788 and 1868, about 162,000 convicts were transported from Britain to Australia. That's 162,000 people who disappeared from British records. And many of them had families, wives, children, siblings, who stayed behind and eventually immigrated to other countries. Second, this is a perfect example of what I call cross-border disappearance research. Your ancestor doesn't have to be a convict for these techniques to work. They just have to be someone who moved from one country's record system to another. I've used this exact same AI workflow to track Germans who immigrated to Brazil, Irish who went to Australia versus America, figuring out which branch went where, Scottish families split between Canada and New Zealand, English ancestors who moved to South Africa during the Boer War era. The methodology is identical. You're using AI to bridge the gap between two different countries' record-keeping systems. And here's the thing that makes AI so powerful for this. These records exist. They're digitized. But they're in different databases, with different search interfaces, using different terminology. Australian records use terms like ticket of leave and conditional pardon. British records call these people transports or convicts. American researchers might not even know those terms exist. But AI? AI can translate between these record systems, understand the terminology, and help you know exactly what to search for and where. So, that's what we're going to do today. We're going to use three completely free AI tools to solve Sarah's mystery about Thomas Bennett. And then I'm going to show you the advanced techniques that take this to the next level. Sound good? Let's do this. 

Okay, so the first thing I did when Sarah sent me this case was to realize I don't know much about the Australian convict system. I mean, I know the basics. Britain sent convicts to Australia. They worked off their sentences. Some stayed. Some returned. But the details, the terminology, the record types, I needed to get up to speed. And this is where perplexity absolutely shines. Because perplexity doesn't just search. It researches with citations. And for historical context, that's exactly what you need. So I opened perplexity, the free version, by the way, which gives you unlimited basic searches. And here's the exact prompt I used. Quote, Research the Australian convict system for convicts transported between 1830 to 1835. I need to understand, one, what records were created when convicts arrived in Australia. Two, what happened after they served their sentence, ticket of leave, conditional pardon, et cetera. Three, what records documented their post-sentence life. Four, where these records are held today and if they're digitized, end quote. why is this prompt effective? Because I'm not asking perplexity to guess or speculate. I'm asking it to research specific factual information. And I'm giving it boundaries, the time period, the specific information I need, and where I want to know if records exist. Within about 30 seconds, perplexity came back with a beautifully organized answer with citations from the National Archives of Australia, the State Library of New South Wales, academic articles about convict record keeping, the Australian government's convict records portal. And here's what I learned. And this is gold for anyone researching transported convicts. Number one, convict indent. When the ship arrived, a detailed document called a convict indent was created listing every prisoner, their crime, their sentence, physical description, literacy level, trade skills, and where they were assigned to work. Number two, tickets of leave. After serving part of their sentence, usually three to four years for a seven-year sentence, convicts could apply for a ticket of leave, essentially parole. These applications are documented and include where they were living, who they were working for, and character references. Number three, certificates of freedom. When the sentence was complete, they received a certificate of freedom, which is documented in colonial records. Number four, post-freedom records. After freedom, convicts appear in regular colonial records, land grants, marriage records, business licenses, court cases, as plaintiffs, defendants, or witnesses, and death records. Number five, return voyages. If a convict wanted to return to Britain, they needed permission and money. Very few actually returned, maybe 10 to 15%, because it was expensive and it built lives in Australia. Perplexity gave me all of this in one search with direct links to where these records are held. I now knew exactly what to look for. Now, here's the advanced technique I want you to know about. Once perplexity gives you that initial research, you can ask follow-up questions to drill deeper. This is where perplexity's conversational nature really shines. Quote, based on that information, if I'm searching for a specific convict named Thomas Bennett who arrived in July 1832 aboard the Florencia, what are the top three most likely record types I should search first and which Australian archives or databases would have those records online? End quote. See what I did there? I took the general knowledge and made it specific to my case. Perplexedly responded with, Number one, convict indent for the Florencia, State Library of New South Wales Digitize. Number two, ticket of leave applications from 1835 to 1837, National Archives searchable. Number three, New South Wales Marriage Index, 1788 to 1888, Ancestry slash Family Search, Free on Family Search. This is the research roadmap. This is AI functioning as your research assistant telling you exactly where to look and in what order. Now, remember our golden rule. AI is your research assistant, not your researcher. Perplexed didn't magically find Thomas Bennett for me. What it did was give me the knowledge I needed to search intelligently instead of just randomly clicking through databases hoping to find something. This is exactly how you should use AI for any international research. When your ancestor crosses a border, you need to understand the new country's record system. Perplexity is the fastest way to get that understanding. 

Okay, so armed with my perplexity research, I now knew where to look. I went to the State Library of New South Wales website and searched their digitized convict indents for the ship Florencia, 1832. And there he was. Thomas Bennett, age 24, convicted of larceny, stealing cloth, seven-year sentence, arrived July 18, 1832. The intent included his physical description, his trade, weaver, and, here's the key detail, where he was assigned. Quote, assigned to Government Road Gang, Blue Mountains, October 1832, end quote. Perfect! Now I knew where he went immediately after arrival. Next step, find his ticket of leave application. I searched the National Archives of Australia's database for Thomas Bennett and ticket of leave between 1835 to 1837, because perplexity told me that's when he'd be eligible. And bingo! There it was. Ticket of leave application approved February 1836. But here's where it gets interesting. The ticket of leave record includes where he was living at the time of the application. Bathurst, New South Wales, that's about 200 kilometers west of Sydney. And it lists his employer, William Lawson, landowner. Now, I've got two new pieces of information, a location and a name. This is where Claude comes in. I downloaded the ticket of leave document. It's a scanned, handwritten form. And I needed to extract every detail from it. This is where Claude's document analysis capabilities really shine. I opened Claude, the free version on Claude. ai, and uploaded the ticket of leave image. Then, I used this prompt. This is a ticket of leave document from 1836 in New South Wales, Australia. 



Claude came back with a beautifully organized summary. Ticket of leave summary. Ticket of leave summary. Applicant, Thomas Bennett. No alias is noted. Crime, larceny. Theft of cloth. Original sentence, seven years transportation. Ship, Florencia, arrived July 1832. Application date, January 15, 1836. Approval date, February 3, 1836. Current resident, Bathurst District, New South Wales. Employer, William Lawson, landowner and magistrate. Occupation, agricultural laborer. Conditions, must remain in Bathurst District unless given written permission. Must attend muster, roll call, monthly. Cannot be on the streets after sunset without permission. Subject to return to assigned labor if violations occur. Character references, William Lawson, employer. Industrious and sober. Reverend Thomas Hassal, local clergyman. Attends church regularly. Approval authority, colonial secretary's office. Official signature, E. D. S. Thomas, colonial secretary. Now, that's useful. But here's where it gets even better. The advanced cloth technique, cross document analysis. I had multiple documents at this point. One, the convict indent, arrival document. Two, the ticket of leave, 1836. Three, a certificate of freedom I'd found, 1839 in the sentence. So, I uploaded all three documents to Claude and used this advance prompt. Quote, I'm uploading three documents about the same person, Thomas Bennett, a convict transported to Australia in 1832. 2. Please analyze all three documents and, 1. Create a chronological timeline of his life from arrival through freedom. 2. Note any discrepancies in information between documents, ages, locations, occupations. 3. Identify any names that appear multiple times across documents. 4. Suggest what type of records I should search next based on the information revealed, end quote. This is where Claude's analytical power really shows. It curated. Timeline. July 1832. Arrived Sydney assigned to Road Gang. October 1832. Transfer to Blue Mountains Road Work. Sometime 1833 to 1835. Transitioned from government work to private employment with William Lawson. January 1836. Applied for ticket to leave. February 1836. Ticket approved. Allowed to work freely in Bathurst District. June 1839. Received Certificate of Freedom. Sentence completed. Discrepancies noted. Age listed as 24 in 1832, but as 29 in 1836. Should be 28. Possible transcription error. Trade listed as weaver,  on arrival but agricultural laborer by 1836. Occupational shift. Recurring names. William Lawson appears in both Ticket of Leave and Certificate of Freedom. Important person. Reverend Thomas Hassall. Character reference. Community connection. Suggested next searches. Number one. Marriage records for Thomas Bennett in Bathurst, 1836 to 1850, now that he had freedom to marry. Number two. Land grant records. Ex-convicts were eligible for land grants after freedom. Number three. Death records for Bathurst District. Number four. William Lawson's property records. Might mention Bennett as employee or tenant. Number five. Church records at Reverend Hassal's church. Baptisms, marriages, burials. Do you see what just happened? Claude didn't just summarize the documents. It reasoned across them. It identified patterns, found discrepancies, and created a research plan. This is AI as your research assistant at its finest. I now had a clear path forward. Now, here's where things get really practical. One of the documents I found, a handwritten letter from Thomas to the colonial secretary in 1838 requesting permission to marry, was almost illegible. Old handwriting. Faded ink. The whole deal. This is where Gemini AI Studio comes in. And this is critical. You need to use Google AI Studio. Aistudio. google.com. Not the Gemini app. AI Studio uses the latest Gemini 3 model, which has near-perfect handwriting transcription. We're talking about 1.67% character error rate. That's expert-level accuracy. I uploaded the handwritten letter image to AI Studio and used this prompt. Quote, Please transcribe this handwritten document exactly as written, preserving original spelling, punctuation, and line breaks. Use brackets for unclear words and question mark for illegible words. This is a formal letter from 1838 to colonial authorities, end quote. Within seconds, Gemini transcribed the entire letter. Here's an excerpt. To His Excellency, the colonial secretary. Sir, I most respectfully beg to submit this petition for permission to marry. I have served my sentence faithfully and have been granted my certificate of freedom. I wish to marry Miss Sarah Mitchell, daughter of John Mitchell, free settler of Bathurst. She is of good character, and I have secured employment with Mr. William Lawson, which will support a family. I remain, sir, your most obedient servant, Thomas Bennett, Bathurst, March 1838. Sarah Mitchell? A name! Now I have something to search. Here's the advanced technique. Once Gemini transcribes the document, ask it to extract structured data. Quote, based on this transcribed letter, create a structured data summary with 1. Sender name and location 2. Recipient and their office 3. Date of letter 4. Primary resident 5. All people mentioned by name with their relationships or roles 6. Any locations or employers mentioned, end quote. Gemini returned Structured data Senator Thomas Bennett, Bathurst, NSW Recipient Colonial Secretary Sidney Date March 1838 Request Permission to Marry People mentioned Sarah Mitchell Intended Bride John Mitchell Sarah's father Described as free settler of Bathurst William Lawson Employer Mentioned as providing ongoing employment. Employment status Currently employed by William Lawson Financial Status Claims ability to support a family Character Emphasizes complete sentence And good behavior Now, I have three more names to research Sarah Mitchell John Mitchell And confirmation that William Lawson Continued to be important in Thomas' life And this is the power of using multiple AI tools in sequence Perplexity gave me the roadmap Claude analyzed and connected the documents I Jim and I made the illegible readable. Each tool doing what it does best. But we're not done yet. Because now comes the verification, and this is where our golden rule really matters. AI is your research assistant, not your researcher. Everything AI has told me so far needs to be verified with actual records. 

Armed with the names from Claude's analysis and Gemini's transcription, I went back to the Australian records databases. I searched for 1. Marriage record, Thomas Bennett and Sarah Mitchell, Bathurst, 1838-1839. Found, marriage entry, June 12, 1838, St. Michael's Church, Bathurst. Reverend Thomas Hassall, officiating, the same reverend from his ticket of leave. Number 2, Sarah Mitchell's family. Found, John Mitchell, free settler, arrived 1830, granted land in Bathurst. Census records showing Sarah Mitchell, born 1818, living with father. Number 3, children of Thomas and Sarah. Found, baptism records. William Bennett, born 1839. Mary Bennett, born 1841. Thomas Bennett Jr., born 1843. Elizabeth Bennett, born 1846. Number 4, land records. Found, land grant to Thomas Bennett, 1845, 50 acres near Bathurst, granted to, free man, former convict. Number 5, death record. Found, Thomas Bennett died, February 1871, Bathurst, age 63, buried at St. Michael's Cemetery. He never went back to England. He built an entire new life in Australia. New wife, four children, land ownership, respected member of the community. The man who was transported as a convict for stealing cloth became a landowner and raised a family in Australia. But here's the truly amazing part. Remember Sarah, the listener who contacted me? Her fourth great-grandfather, Thomas Bennett, the one who disappeared from England in 1831? He left behind a wife, Mary Clark, and a daughter, Elizabeth, in London. That Elizabeth grew up, married, and eventually immigrated to America in the 1860s. That's Sarah's line. The American line from Thomas' first family. But Thomas' second family, his Australian children, they stayed in Australia. Which means Sarah has Australian cousins she never knew existed. Four entire family lines descended from Thomas' Australian children. Now, let me bring this back to you and your ancestors. Because this exact research pattern works regardless of whether you have Australian convicts. The methodology is... Step 1. Use perplexity to understand the record system of the destination country. What records were created? Where are they held? What's the terminology? What's digitized? Step 2. Use Cloud to analyze and connect documents you find. Look for patterns. Identify discrepancies. Create timelines. Generate next-step research plans. Step 3. Use Gemini AI Studio to transcribe any handwritten documents. Make the illegible readable. Extract structured data. Find names and dates for further researching. Step 4. Verify everything with actual records. Never trust AI output without verification. Use AI to guide you to the right records. Confirm with primary sources. This works for... Germans in Brazil or Argentina. Irish in America vs. Australia. Italians in America vs. South America. Scots in Canada vs. New Zealand. Any cross-border migration pattern. The AI tools are the same. The record types differ, but the methodology is identical. Here's why this episode matters, even if you're certain you don't have Australian ancestors. About 20% of Americans have British or Irish ancestry. And during the peak transportation period, 1788-1868, about 162,000 people were transported from Britain to Australia. That means there's a statistical likelihood that some percentage of American genealogists researching British lines will encounter a transported convict somewhere in their tree. But most people don't find them because they stop searching when the ancestor disappears from British records. Now you Now you know. If your British or Irish ancestor vanishes between 1788 and 1868, check the Australian convict databases. It's free. It's digitized. And the research techniques I just taught you will work. Plus, and this is huge, many British families were split by transportation. One sibling was transported, the rest stayed in Britain, and some of those eventually immigrated to America. You might have Australian cousins you don't know about, descended from the transported family member. DNA testing is revealing these connections every single day. People are getting DNA matches to Australian cousins and having no idea how they're related. Now you know how to find that connection. 

Okay, let's recap exactly what we did today so you can use these techniques yourself. Tool number one, perplexity free. Used to research the Australian convict system. Got the roadmap, what records exist, where they're held, what terminology you use. Created a research plan based on factual information with citations. Tool number two, Claude, free. Analyzed multiple documents to create timelines. Identified patterns and discrepancies. Generated next-step research suggestions and connected information across sources. Tool number three, Gemini AI Studio, free. Transcribed handwritten documents with near-perfect accuracy. Extracted structured data from transcriptions. Made illegible documents readable. Now, here's the advanced technique I want to mention. After I finished all this research, after I had the documents, the verified timeline, the family connections, I wanted to create something Sarah could share with her family. So I uploaded all of my research documents to Google Notebook LM. The convict indent, the convict indent, the ticket of leave, the marriage record, the land grant, my research notes. And I asked Notebook LM to create an infographic showing Thomas Bennett's life journey from convict to landowner. It produced a beautiful visual timeline that Sarah could share with her family, showing his transportation 

his service and ticket of leave. His service and ticket of leave, his marriage and family, his land grant, his death in 1871. Notebook LM can also create audio overviews like mini-podcasts about your ancestor, slide decks for family presentations, and detailed summaries of your research. This is the next-level workflow. Use Perplexity, Claude, and Gemini to do the research and analysis. Then use Notebook LM to create shareable visual summaries for your family. All of these tools are free, every single one. 

Here's your homework assignment for this week, and I want you to share your results in our Facebook group using the hashtag #MyInternationalAncestor. Step 1, pick one ancestor who crossed an international border. Doesn't have to be Australia. Could be Italy to America, Ireland to Canada, Germany to Brazil, anywhere. Step 2, open Perplexity and use this prompt. Quote, research the immigration/emigration records for people traveling from origin country to destination country between years. I need to understand, one, what records documented their departure? Two, what records documented their arrival? Three, what records were created after their arrival? Four, where these records are held today and if they're digitized? End quote. Step 3, take a screenshot of Perplexity's answer and post it in the Facebook group with the hashtag #MyInternationalAncestor. Tell us, what new record type did you learn about that you didn't know existed? That's it. You don't have to solve the whole mystery this week. Just use Perplexity to understand the record system. That's your research foundation. Thank you so much for listening to Ancestors and Algorithms. If you enjoyed this episode and especially if you learned something new about international genealogy research, please leave a review wherever you listen to podcasts. Reviews help other family historians find us and every single one means the world to me. Don't forget, join our Facebook group, Ancestors and Algorithms AI for Genealogy. That's where over 1,300 genealogists are sharing their AI breakthroughs, asking and helping each other solve brick walls. It's an incredible community and I'd love to see you there. Before I let you go, let me leave you with this. These tools are incredible for understanding record systems, analyzing documents, and creating research plans. But you're still the genealogist. You're still the one who verifies with primary sources, applies proof standards, and makes the final decisions. AI helps you work smarter. You bring the expertise. I'm your host, Brian, and I will see you next week for another journey into the past powered by the future. Until next time, happy researching.