Guwahati: Can artificial intelligence help preserve a language that is slowly disappearing, decipher a damaged manuscript or record an oral tradition before it is lost to time?

Researchers led by Nagaland University believe AI could play an important role in safeguarding India’s cultural heritage, but caution that technology must not come at the cost of cultural accuracy, community ownership and traditional knowledge.

A multi-institutional research team has proposed a framework to evaluate the use of AI in preserving manuscripts, endangered languages, oral traditions, historical records and other forms of cultural heritage.

The study, published in the peer-reviewed journal Acta Innovations, argues that digitising historical material is not enough. AI tools must also be able to retain the historical meanings, cultural references and social contexts embedded in the material.

The research brings together scholars from seven institutions across Nagaland, Tamil Nadu, Delhi, Uttar Pradesh, Telangana and Karnataka.

The authors include Ravi Ranjan Kumar of Nagaland University, Major N Kavitha of Vellalar College for Women, Nisha Sengar of PGDAV College (Evening), University of Delhi, Sonia Gouri of ABES Engineering College, J Lalitha Sridevi of Hyderabad, Kumar Sai Deepak B of Sapthagiri NPS University, Bengaluru, and P Deepavathi of Joy University, Tamil Nadu.

Nagaland University Vice-Chancellor Jagadish K Patnaik said the collaboration reflected the potential of AI to help document and transmit India’s cultural knowledge to future generations.

He, however, stressed that technological interventions must respect cultural contexts, community rights and ethical considerations.

“A key contribution of our study is a framework that evaluates AI-assisted cultural preservation across nine areas,” said Ravi Ranjan Kumar of the Hindi Department, Nagaland University.

These include digitisation accuracy, text recognition, classification, translation, visual reconstruction, descriptive information, cultural context, accessibility and ethics.

The framework combines technical assessments with reviews by subject experts and the communities associated with the material.

Kumar pointed out that technical accuracy did not necessarily guarantee cultural accuracy.

An AI system might successfully transcribe a document or translate spoken words but still miss historical references, cultural nuances or the significance of a particular tradition.

This becomes particularly important when dealing with less commonly used languages, unfamiliar scripts, damaged manuscripts and oral histories.

The researchers recommend that AI should assist, rather than replace, archivists, linguists, historians and cultural custodians. Community participation, they argue, is essential when documenting traditions and deciding how cultural material can be accessed and used.

Applications beyond digitisation

The proposed framework identifies five broad areas where AI could support heritage preservation: printed and handwritten documents, photographs and artworks, oral histories and endangered languages, archaeological and architectural heritage, and folklore and community knowledge.

Potential applications include making old manuscripts searchable, organising historical photographs, identifying damage to artworks, transcribing oral recordings, translating multilingual material and developing digital maps and three-dimensional models of heritage sites.

The study also draws attention to the difference between restoring existing material and reconstructing information that has been lost.

The researchers say AI-generated interpretations must be clearly identified, with original digital files retained and changes documented. This would help prevent computer-generated reconstructions from being mistaken for historical evidence.

The framework could have particular relevance in Northeast India, where numerous indigenous languages, oral traditions, folklore and community histories remain to be documented.

For communities across Nagaland and other northeastern states, AI-assisted documentation could offer new ways of recording languages and preserving oral histories. However, the researchers caution that such efforts must involve the communities concerned and respect their rights over cultural knowledge.

The study also flags concerns around data privacy, access and the accuracy of AI-generated translations and descriptions.

For universities, museums, libraries and archives, searchable digital collections could make historical material more accessible to students and researchers, while reducing the need to repeatedly handle fragile original documents.

The researchers have called for the framework to be tested across different collections, scripts and dialects, with expert verification and community feedback forming part of the assessment.

The larger question, the study suggests, is not simply how much of India’s cultural heritage can be digitised using AI, but whether the technology can preserve the meaning and identity behind it.