A recent study describes a new AI that produces high-quality translations of ancient texts, specifically cuneiform tablets. Translating an ancient language is a time-intensive process, and only a few hundred experts are qualified to perform it. The AI model was trained on a sample of cuneiform texts and learned to translate Akkadian from transliterations and cuneiform symbols directly. The model was able to reproduce the nuances of each test sentence’s genre, showing promise for potential uses beyond translation. Although there is still work to be done to improve accuracy, the researchers hope that the AI model can assist human scholars in translating and preserving the cultural heritage of ancient Mesopotamia.
| Signal | Change | 10y horizon | Driving force |
|---|---|---|---|
| New AI translates 5,000-year-old cuneiform tablets instantly | Translation process from ancient to modern language | More efficient and accurate AI translations | Preservation and dissemination of cultural heritage |
| Few experts qualified to translate ancient languages | Limited number of translators for ancient languages | Increased accessibility to translated ancient texts | Need to unlock knowledge and information from ancient texts |
| AI model translates Akkadian cuneiform | AI translates ancient Akkadian language | Virtual assistant aiding human scholars in translations | Preservation and dissemination of cultural heritage |
| Cuneiform tablets preserve ancient history | Cuneiform tablets provide historical records | Increased understanding of ancient civilizations | Preservation of cultural heritage |