Google’s Aeneas AI helps historians decode ancient Latin inscriptions

Google DeepMind has launched Aeneas, an AI system designed to help historians decode and contextualize ancient Latin inscriptions carved in stone. The tool analyzes weathered engravings to determine when and where they were originally created, while also providing researchers with historical parallels from a database of nearly 150,000 catalogued inscriptions spanning from modern-day Britain to Iraq.

How it works: Aeneas processes partial transcriptions alongside scanned images of inscriptions to reconstruct missing text and provide historical context.

• The system can fill in damaged portions of text—for example, completing “…us populusque Romanus” by suggesting “Senat” to form “Senatus populusque Romanus” (“The Senate and the people of Rome”).
• Unlike its predecessor Ithaca (which focused on Greek texts), Aeneas cross-references inscriptions with its stored database to identify similar engravings that feature comparable words, phrases, and analogies.
• The AI was trained on approximately 150,000 inscriptions and several thousand images, a relatively small dataset compared to the billions of documents used for general-purpose language models like Google’s Gemini.

Why specialized tools matter: The limited availability of high-quality inscription scans necessitates purpose-built solutions rather than general AI models.

• “There simply aren’t enough high-quality scans of inscriptions to train a language model to learn this kind of task,” the researchers noted.
• Rather than automating epigraphy entirely, the team aims to create “a tool that will integrate with the workflow of a historian,” according to Yannis Assael, a Google DeepMind researcher on the project.

Validation results: Testing with 23 historians showed significant improvements in research workflows and accuracy.

• The study, published in Nature, found that Aeneas sparked research ideas for 90% of inscriptions and led to more accurate determinations of origin dates and locations.
• When tested on the Monumentum Ancyranum—a famous inscription in Ankara, Turkey—Aeneas provided estimates and parallels that matched existing historical analysis with attention to detail resembling a trained historian’s approach.

What experts are saying: Initial reactions from the research community have been enthusiastic, though questions remain about real-world applications.

• “That was jaw-dropping,” said Thea Sommerschield, an epigrapher at the University of Nottingham who worked on Aeneas.
• However, Kathleen Coleman, a Harvard classics professor, noted that the tool doesn’t interpret text meanings and its long-term utility remains unclear, especially given that testing focused on well-studied inscriptions rather than obscure samples.

Accessibility and future applications: Google DeepMind has made Aeneas open-source and freely available to educators, students, museum workers, and academics.

• The team is collaborating with Belgian schools to integrate the tool into secondary history education.
• Sommerschield envisions field applications: “To have Aeneas at your side while you’re in the museum or at the archaeological site where a new inscription has just been found—that is our sort of dream scenario.”