The Ministry of Electronics and Information Technology (MeitY) has launched a Handbook based on the Data Empowerment and Protection Architecture (DEPA) for AI training. Its objective is to promote secure and trusted data collaboration while safeguarding privacy, data rights and institutional control.
During a high-level roundtable conference held in Bengaluru, MeitY Secretary S. Krishnan unveiled this handbook in collaboration with the iSPIRT Foundation.
The framework aims to make it easier for organisations to share and use high-quality data for AI training without compromising privacy or losing control over their information. It brings together legal agreements, secure computing systems, and privacy protection measures to enable the safe, collaborative use of data. This will also help track the origin of the data, the permissions that apply to it, and how it is used during the AI training process.
Explaining the functioning of this framework, Prof. Ajay Kumar Sood, Principal Scientific Adviser to the Government of India, stated that instead of bringing data to the AI model, the model is taken to the data. He added that this would foster secure and trusted collaboration among various institutions and open up new opportunities for cross-border research partnerships.
Anjani Rathor, Group Head of Digital Banking and Customer Experience at HDFC Bank, highlighted opportunities to leverage the bank’s experience with DEPA (Data Empowerment and Protection Architecture) through DEPA training. Sharad Sharma, co-founder of the iSPIRT Foundation, described this initiative as a significant milestone in India’s evolving data economy.
Krishnan said secure and responsible access to data would be crucial for India’s AI goals, while Infosys co-founder Kris Gopalakrishnan highlighted the framework potential to strengthen research partnership and help develop more reliable and advance AI models.
Representatives from the government, industry, technology, academia and research sectors participated in this roundtable conference. The objective of this initiative is to facilitate access to various datasets to foster AI innovation, while protecting the interests of data custodians and contributors.
(PIB)
