AI in Genealogy: Latest Innovations No. 1

From Manuscripts to Memories: Cutting-Edge AI in Historical Research

Smart Dating of Ancient Texts
A groundbreaking tool called Enoch is transforming the way historians date handwritten manuscripts. By blending handwriting-style features with radiocarbon-dated data and leveraging Bayesian ridge regression, Enoch can estimate manuscript dates—like the Dead Sea Scrolls—with an average error of just 28–31 years. This offers more nuanced chronological insights than traditional paleographic methods.

AI as Digital Legacy: “AI Afterlives” & Ethical Doppelgängers
Recent academic research delves into the complex world of using AI-generated agents as digital continuations of ourselves—termed “AI Afterlives.” These models emulate an individual’s persona and raise important questions around identity, privacy, consent, and interactions with future generations.

Surnames & Systemic Bias
An intriguing new study reveals that AI decision-making systems may exhibit bias based on surnames—favoring “elite” names by associating them with competence, intelligence, or wealth. This raises concerns for genealogical tools, potentially reinforcing privilege through algorithmic design.


Popović, Mladen, Maruf A. Dhali, Lambert Schomaker, Johannes van der Plicht, Kaare Lund Rasmussen, Jacopo La Nasa, Ilaria Degano, Maria Perla Colombini, and Eibert Tigchelaar. “Dating Ancient Manuscripts Using Radiocarbon and AI-Based Writing Style Analysis.” PLOS One 20, no. 6 (2025): e0323185. Accessed September 12, 2025. https://doi.org/10.1371/journal.pone.0323185. PLOS

Lei, Ying; Shuai Ma; Yuling Sun; Xiaojuan Ma. “‘AI Afterlife’ as Digital Legacy: Perceptions, Expectations, and Concerns.” Preprint, arXiv, February 15, 2025. Accessed September 12, 2025. https://arxiv.org/abs/2502.10924. arXiv

Pataranutaporn, P., Nattavudh Powdthavee, Pattie Maes. “Algorithmic Inheritance: Surname Bias in AI Decisions Reinforces Intergenerational Inequality.” Preprint, arXiv, January 23, 2025. Accessed September 12, 2025. https://arxiv.org/abs/2501.19407. arXiv

Leave a Reply