These tools can be a good support, but they're not a substitute for humans. To present them as such does people using them a disservice. I would challenge genealogists using them to pay attention to what they can and cannot do. #genealogy #AI
05.03.2026 14:28 β
π 1
π 0
π¬ 0
π 0
2) Historical context doesn't just mean "I looked at a dozen reference books and told you what they said." That's not how #history works as a field. Instead, building historical context requires building an analysis based on what's known and unknown.
05.03.2026 14:26 β
π 2
π 0
π¬ 0
π 1
1) Machine translation is not 100% reliable. In many fields, "close enough" #translation is considered acceptable. In #genealogy, "close enough" can mean the wrong death date. Are we okay with that as researchers?
05.03.2026 14:25 β
π 2
π 0
π¬ 0
π 0
HTR and OCR are similar mechanically, but with the fact that HTR needs to account for a multitude more variables, the risk of error is much higher. And HTR being used directly for machine translation should not be happening...
03.03.2026 17:07 β
π 1
π 0
π¬ 0
π 0
It's essentially taking pieces of prior translations used in its training. Machine translation functions like this generally, but we were often not using machine translation in genealogy because of the handwriting. That's changed.
03.03.2026 16:49 β
π 1
π 0
π¬ 1
π 0
I've been explaining it as pattern recognition. Which can become problematic in translation because the LLM will translate the expected pattern, sometimes instead of the actual document. It looks good so, it will often get used as the basis for further research. That can prove problematic later.
03.03.2026 16:43 β
π 1
π 0
π¬ 1
π 0
I wish we could break the pattern of thinking it's okay not to expect genealogists to learn and/or teach genealogists about other fields. I suspect it's a large part of the problem. If you don't understand methodology, it's easier to accept the outputs.
03.03.2026 11:46 β
π 1
π 0
π¬ 1
π 0
In terms of #genealogy, I narrowed the question in large part because professionals seem to be discouraging the use of #AI for the use of research. But that is not happening for #translation. And, honestly, that means that some clients are research reports created based on wrong translation...
03.03.2026 11:44 β
π 0
π 0
π¬ 1
π 0
Okay, at some point is #genealogy going to have a conversation about the ethics of #AI #translation?
It's being used by genealogists, including professionals. There are implications. Discussions need to happen.
02.03.2026 21:11 β
π 1
π 1
π¬ 2
π 0
My family's surprise goldmine were US passport applications, as they required the listing of the father's place of birth if a woman was unmarried.
02.03.2026 14:11 β
π 0
π 0
π¬ 0
π 0
How accurate was it? : Some major errors risk destroying your research if you rely on it going forward. 1) Marguerite's surname is Bigue, not Richard. 2) the death date was the thirty-first of the preceding month. (3: less important: most of the witness information is transcribed wrong.)
01.03.2026 13:12 β
π 3
π 0
π¬ 0
π 0
veuf de Marguerite Ricard de la paroisse de
St. Anne dΓ©cΓ©dΓ© le premier du mois prΓ©cΓ©-
-dent, Γ’gΓ© de quatre-vingt six ans, en prΓ©-
-sence de soussignΓ©s Louis Lafleche & HonorΓ©
Dion qui nβont sΓ§u signer, dont lecture faite.
[SignΓ©] Louis Lafleche
[SignΓ©] HonorΓ© Dion + sa marque
[SignΓ©] M. Chauvin Ptre
01.03.2026 13:12 β
π 0
π 0
π¬ 0
π 0
[Main Body]
Lβan mil huit cent quarante, le trois
septembre, Nous soussignΓ© CurΓ© de St. Anne
avons inhumé dans le Cimetière de cette
paroisse le corps de Modeste Richer Lafleche
01.03.2026 13:12 β
π 0
π 0
π¬ 1
π 0
Transcription: SΓ©pulture de Modeste Richer Lafleche
[Left Margin]
SΓ©p: 3e
Modeste
Richer
Lafleche
01.03.2026 13:11 β
π 0
π 0
π¬ 1
π 0
Below is the diplomatic transcription, preserving the original spelling, punctuation, and line breaks as written by the priest, M. Chauvin.
01.03.2026 13:11 β
π 0
π 0
π¬ 1
π 0
Response:
This record for Modeste Richer Lafleche is found on the left-hand page of the register from the parish of Sainte-Anne-de-la-PΓ©rade, dated September 1840.
01.03.2026 13:11 β
π 0
π 0
π¬ 1
π 0
Prompt: Please transcribe using diplomatic translation the burial of Modeste Richer LaFleche.
01.03.2026 13:11 β
π 0
π 0
π¬ 1
π 0
Genealogists interested in #AI have touted the transcription ability of Google Gemini's paid edition. So how well does it actually work? #genealogy #translation #French #familyhistory
01.03.2026 13:11 β
π 4
π 1
π¬ 2
π 0
Looking for help understanding legal terms? This glossary covers some of the more unusual terms. #genealogy #ancestry #French #translation
28.02.2026 23:09 β
π 0
π 0
π¬ 0
π 0
ATAβs Mastermind Program offers mentoring designed to help you learn and develop professionally. Apply for the Mastermind Program at https://bit.ly/479t9RB! The deadline to apply is February 28--Don't go it alone!
.
.
.
#Career #Mentoring #Translator #Interpreter #xl8 #1nt #atanet
28.02.2026 17:05 β
π 0
π 2
π¬ 0
π 0
π’ ATA is excited to accept session proposals for ATAβs 67th Annual Conference, October 28-31, 2026, in San Francisco, California! Submit your proposal today at https://bit.ly/4kmhhBr! π
.
.
.
#Career #Growth #Translator #Interpreter #Experience #LanguageServices #PublicSpeaking #xl8 #1nt #atanet
28.02.2026 20:57 β
π 3
π 3
π¬ 0
π 1
Graphic for the Association of Professional Genealogists with the APG logo and βExpert Insights.β A magnifying glass highlights the word βprofessionalβ in a dictionary. Promotes βHow/When to Hire a Proβ with Michelle Dwyer Cohen, Friday at 10:00 AM and 1:30 PM, live at RootsTech booth 1135.
Stuck on a research problem? Learn how and when to hire a professional genealogist with Michelle Dwyer Cohen on Friday 6 March at 10:00 AM and 1:30 PM at the APG RootsTech booth #1135. #RootsTech2026 #APG
28.02.2026 19:00 β
π 2
π 2
π¬ 0
π 0
2026 PMC Call for Proposals - Association of Professional Genealogists
I would love to see a #genealogy #translator who isn't me apply to teach on #translation in genealogy. There's a need for new voices! www.apgen.org/2026_pmc_cal...
27.02.2026 20:59 β
π 0
π 0
π¬ 0
π 0
I wish more genealogists understood the principle of reproduction rights...
23.02.2026 21:53 β
π 1
π 0
π¬ 0
π 0
I'm more protective of mine, but many of the ones I'm handling are from the family studio, and we've had past issues with organizations trying to claim rights they don't own.
23.02.2026 21:53 β
π 1
π 0
π¬ 1
π 0