Generative AI is my current obsession. I find it both useful and intriguing. While it could be a bit of a challenge getting on the platform in the beginning, once I subscribed, it was no problem, and I use ChatGPT every day for a variety of purposes.

One thing I hadn’t tried ChatGPT for was genealogy research assistance. As far as AI, we all have most likely used it indirectly. The 1950 U.S. census was indexed using AI when it was released last year. A few years ago, during RootsTech, my favorite thing was the AI trial that ran during the conference allowing users to transcribe documents. While everyone else was attending presentations, I spent nearly every day scanning and transcribing. That week, the 1841-1860 Washington Presbyterian Church Minutes, Guernsey County, Ohio was transcribed. Did it have some errors? Yes. Did I fix them? Yes. Did it save me weeks of time? Absolutely.

So, how could AI help with genealogy now? Well, I asked Chat GPT.

ChatGPT AI for Genealogy Research

ChatGPT is very easy to use. You simply type a question into the box, and it gives you an answer. If you want to see what else it can find, click regenerate response.

Using a combination of two responses, I ended up with:

AI can be a valuable tool for genealogy research, particularly in assisting with tasks such as record indexing, transcription, and analysis of DNA data. Here are some ways people can use AI for genealogy research:

• Record indexing: AI can help index large collections of genealogy records, making it easier for researchers to search for specific ancestors. For example, the FamilySearch website uses AI to index millions of historical records, such as census and vital records, making them searchable by name, location, and other criteria.
• Document analysis: AI can also analyze historical documents such as census records, birth and death certificates, and immigration records. AI-powered tools like DocAI and Amazon Textract can extract information from these documents and provide insights into family relationships, locations, and occupations.
• Transcription: AI can transcribe handwritten or difficult-to-read documents, such as census records or wills. Using machine learning algorithms, AI can learn to recognize different handwriting styles and even decipher old or faded text.
• Family tree construction: AI-powered tools like FamilySearch and Ancestry.com can help construct family trees by analyzing historical records and DNA data. These tools use machine learning algorithms to suggest possible family connections and help users build their family trees.
• DNA analysis: With the increasing popularity of DNA testing for genealogy purposes, AI can analyze DNA data and identify potential matches between individuals. For example, AncestryDNA uses AI to analyze DNA data from millions of customers to help identify potential relatives and build family trees.
• Image recognition: AI can also be used to help identify individuals in old family photographs. By using facial recognition algorithms, AI can compare facial features in photos and help identify potential matches.

Overall, AI can help genealogy researchers process and analyze large amounts of data more quickly and accurately, leading to discoveries and breakthroughs in family history research.

So, quite a few of the responses were about how Family Search and Ancestry are using AI, but there were a few ideas about how individuals could use ChatGPT on their own.

Asking ChatGPT for its sources

If you are like me, you wonder how ChatGPT came up with the answers. I frequently ask for the sources after I submit a question. When asked for sources for this search, ChatGPT responded:

Me: What were the sources for that answer?

ChatGPT: As an AI language model, my responses are generated based on the vast amount of text data that I have been trained on, including articles, books, and other sources of information. In this case, my response on how people can use AI for genealogy research was based on my understanding of the applications and use cases of AI in the field of genealogy research, as well as my knowledge of the capabilities of AI technologies such as record indexing, transcription, DNA analysis, and image recognition. My answer is based on my general understanding of the topic as a language model, and I did not use any specific sources for this answer. However, I can provide references and resources if you would like to explore the topic further.

Well, then, Me: Can you tell me what are some references and resources?

ChatGPT: Certainly! Here are some references and resources you might find helpful for learning more about the ways AI can be used for genealogy research:

1. “How Artificial Intelligence Can Help Genealogy Research,” by Kimberly Powell on ThoughtCo. This article provides a good overview of how AI can be used for genealogy research, including record indexing, transcription, DNA analysis, and image recognition.
2. “The Future of Genealogy and Artificial Intelligence,” by Dr. Maurice Gleeson on FamilyTreeDNA. This blog post discusses the potential impact of AI on the future of genealogy research, including the use of machine learning algorithms for DNA analysis and the integration of AI technologies into genealogy software.
3. “Using Machine Learning to Transcribe Handwritten Documents,” by Ben Lee on Ancestry Tech Blog. This blog post explains how machine learning algorithms can be used to transcribe handwritten documents, and provides examples of how this technology is being used by Ancestry.com to transcribe census records.
4. “The Intersection of Artificial Intelligence and Genealogy,” by John D. Reid on Canada’s Anglo-Celtic Connections. This blog post explores some of the challenges and opportunities presented by the use of AI in genealogy research, and provides examples of how AI is being used by companies such as MyHeritage and AncestryDNA.

These resources should provide you with a good starting point for exploring the ways AI can be used for genealogy research, and may lead you to additional sources and information as well.

Checking for plagiarism and accuracy

Even though no sources were cited in the response, I wanted to see if anything was plagiarized in the original question’s response. So, I popped over to Grammarly to check for plagiarism (it requires a paid subscription).

In the last sentence, the words “analyze large amounts of data more quickly and accurately,” came from an IBM article titled, “What is deep learning?”

“Overall, AI can help genealogy researchers process and analyze large amounts of data more quickly and accurately, leading to new discoveries and breakthroughs in family history research.”

Oddly, ChatGPT frequently answers in a passive voice and may contain grammatical or factual errors, so check all content thoroughly.