Under the proposal, "Training Data Providers" such as hospitals, banks and government agencies would share anonymized and encrypted data with a "Data Exchange Platform". "Training Data Consumers" would then access that data to train their models, inside what MeitY calls hardware-based Secure Data Environments where "data remains encrypted and isolated even during processing". Providers set the rules for how their data can be used, differential privacy techniques are meant to stop models revealing information about individuals, and all activity is immutably logged for audit.

The stated motive is to create AI models trained on Indian data, on the view that offshore model-makers will not be motivated to address the nation's problems or keep data owners in control. MeitY's ambitions are broad: unlocking fragmented health data for diagnostics and drug discovery, extending formal credit to a claimed 400 million underserved individuals and small businesses through shared financial data, and enabling personalised learning for India's 250 million students.

India already operates several data exchanges, but it is unclear how much work remains to make DEPA a reality. The framework also frames the scheme internationally, calling it India's contribution to global science diplomacy and tying it to the country's digital public infrastructure approach, in which government-funded systems such as the Unified Payments Interface are built to limit Big Tech dominance and gifted to friendly nations.