Lumami: A multi-university research team led by Nagaland University has proposed a framework to evaluate how artificial intelligence (AI) can help preserve India’s cultural heritage, including ancient manuscripts, endangered languages, oral traditions and historical records, while safeguarding cultural contexts and community rights.
The study brings together researchers from universities and academic institutions across Tamil Nadu, Delhi, Uttar Pradesh, Telangana and Karnataka.
It examines how emerging AI technologies can support the documentation, preservation and accessibility of cultural materials that face the risk of deterioration, damage or loss as traditional knowledge systems gradually fade.
Moving beyond the digitisation of historical records, the researchers argue that AI-assisted preservation must also account for historical accuracy, cultural meanings and the communities associated with the heritage being documented.
They emphasise that technological interventions should complement, rather than replace, the work of archivists, linguists, historians and cultural custodians.
The findings, published in the peer-reviewed, open-access journal Acta Innovations, propose a framework to assess AI applications across nine parameters: digitisation accuracy, text recognition, classification, translation, visual reconstruction, descriptive information, cultural context, accessibility and ethics.
The framework combines technical assessments with evaluations by subject experts and relevant communities to determine whether AI-generated outputs are not only technically accurate but also culturally appropriate.
Prof. Jagadish K Patnaik, Vice-Chancellor of Nagaland University, said the collaborative research reflected the growing potential of artificial intelligence in safeguarding India’s diverse cultural heritage.
He said integrating AI into the preservation of manuscripts, endangered languages, oral traditions and historical records could open new avenues for documenting and transmitting valuable cultural knowledge to future generations.
However, Patnaik stressed that technological interventions must respect cultural contexts, community rights and ethical considerations. He also acknowledged the contributions of researchers from different institutions towards developing an inclusive and culturally sensitive framework for heritage preservation.
Prof. Ravi Ranjan Kumar of the Hindi Department at Nagaland University, who led the research, said a key contribution of the study was its attempt to establish measurable parameters for evaluating AI-assisted preservation.
He cautioned that technical accuracy does not necessarily translate into cultural accuracy. An AI system may successfully transcribe a historical document or translate spoken words but still fail to capture historical references, cultural nuances or the significance of a particular tradition.
Expert verification and community consultation are therefore essential, particularly when dealing with less commonly used languages, unfamiliar scripts, damaged manuscripts and oral histories, Kumar said.
The study identifies several potential applications of AI in cultural preservation, including making historical documents searchable, organising photographs and artwork collections, detecting damage to heritage objects, transcribing oral recordings, translating multilingual materials and developing digital maps and three-dimensional models of heritage sites.
It also examines the use of AI in preserving archaeological and architectural heritage, folklore and community knowledge, alongside printed and handwritten documents, photographs, oral histories and endangered languages.
An important distinction highlighted by the researchers is the difference between restoration and reconstruction. While restoration seeks to preserve or recover existing material, reconstruction may involve using computational methods to infer information that is missing or unavailable.
The authors recommend retaining original digital records, documenting every AI-generated alteration and clearly identifying areas of uncertainty. This, they said, would help researchers and future users distinguish original historical evidence from computer-generated interpretations.
The framework also calls for safeguards relating to data ownership, privacy, access and community participation. The researchers argue that communities whose languages, traditions and knowledge systems are being documented must have a say in how the material is recorded, interpreted, stored and shared.
The proposed approach could have particular relevance for India’s Northeast, where several indigenous languages, oral traditions and community knowledge systems face challenges relating to documentation and intergenerational transmission.
AI-assisted tools could potentially help document oral histories, record lesser-used languages and make historical collections more accessible. However, the researchers caution that such applications must be evaluated for their suitability to individual languages and dialects, with communities actively involved in decisions concerning their cultural knowledge.
For libraries, museums, universities and archives, searchable digital collections could improve access for students and researchers while reducing the need to repeatedly handle fragile original documents. Multilingual digital access could also bring greater visibility to historical collections that have received limited academic attention.
Despite the potential applications, the study acknowledges several challenges that require further investigation. These include the reliability of AI tools when processing damaged manuscripts, their ability to recognise less commonly used languages and scripts, and the extent to which communities can validate AI-generated translations and descriptions.
The authors recommend piloting the framework across different collections, scripts and dialects to assess its practical effectiveness. Such evaluations should document recognition errors, incorporate feedback from experts and communities, and establish whether AI-generated outputs are suitable for their intended purposes.
The long-term preservation of digital records is another concern, particularly as technologies, software and storage systems evolve.
The research was co-authored by Prof. Ravi Ranjan Kumar of Nagaland University; Major Dr N. Kavitha of Vellalar College for Women (Autonomous), Tamil Nadu; Nisha Sengar of PGDAV College (Evening), University of Delhi; Dr Sonia Gouri of ABES Engineering College, Ghaziabad; Dr J. Lalitha Sridevi of Hyderabad; Kumar Sai Deepak B of Sapthagiri NPS University, Bengaluru; and Dr P. Deepavathi of Joy University, Tamil Nadu.
The research paper is published in Acta Innovations.
The researchers maintain that the future of AI-assisted cultural preservation will depend not simply on technological capabilities but on how effectively innovation is balanced with historical integrity, ethical responsibility and the rights of communities whose heritage is being preserved.
Also Read: Beyond illness: Why true healing begins with listening to people
You just read a story that took days to report. Help us keep our reporters on the ground in the Northeast.
Ad-free reading, support and keep important stories alive
Support once (any amount)
Scan to pay via UPI