Archive · 18 Sep 2026
India Adviser Warns Data Localisation Could Raise Costs, Trigger Retaliation
A government paper is discussing how India should handle data used by artificial intelligence.
It warns that forcing all data to stay in India could make it more expensive for small companies to follow the rules.
It could also stop researchers from using useful data from other countries.
Other countries might respond with trade restrictions.
However, sending data freely abroad can create risks involving foreign surveillance, weak protections, and powerful technology companies controlling too much data.
The paper suggests treating data differently depending on how sensitive it is.
Ordinary statistics could move more freely, while sensitive or national-security data would face tighter controls.
It also recommends creating a group to help regulators coordinate AI data rules.
The paper says its ideas are for discussion and are not official government policy.
A government discussion paper warns that mandatory domestic data storage and processing could raise compliance costs for startups and MSMEs.
The paper says strict localisation could limit access to global datasets, fragment AI research, and provoke retaliatory trade measures.
It proposes a four-tier, risk-based framework allowing different cross-border pathways for aggregated, identifiable, strategic, and national-security data.
The paper recommends an interdisciplinary AI-data interoperability body to coordinate with existing regulators rather than replace them.
The document is a discussion paper, not a final policy position, and notes that its economic estimates are based on hypothetical scenarios.
- Who
- The Office of the Principal Scientific Adviser to the Government of India produced the discussion paper, with contributions from government and industry experts.
- What
- The paper examines cross-border data transfers for AI and proposes a risk-based interoperability framework and a coordinating body.
- When
- The paper was made public on Thursday; relevant Digital Personal Data Protection Act obligations are scheduled to begin on 13 May 2027.
- Where
- The proposals concern India’s data-governance system and the international movement of data.
- Why
- The paper seeks to balance AI research and digital trade with privacy, national security, regulatory oversight, and foreign-surveillance risks.
Localisation and Control
Cross-Border Interoperability
Economic and research effects
Localisation and Control
Keeping data and processing within India can support domestic control and reduce exposure to foreign jurisdictions, but it may increase compliance costs.
Cross-Border Interoperability
More open, compatible data flows can support AI research, startups, services exports, and access to diverse datasets.
Privacy and security
Localisation and Control
Unrestricted international transfers could expose data to foreign surveillance, weaker safeguards, and concentration among major technology firms.
Cross-Border Interoperability
Risk-based transfers, controlled-access agreements, and secure model-to-data processing can enable useful international cooperation without allowing all data to move freely.
Trade and regulation
Localisation and Control
Domestic storage requirements provide stronger sovereign oversight and are already used in some sensitive sectors, such as payment data.
Cross-Border Interoperability
Highly restrictive or fragmented rules could trigger retaliation and hinder digital trade, so the paper favors coordination with existing regulators rather than blanket restrictions.
- Paper title
- Enabling Cross-Border Data Interoperability for AI Systems
- Publishing office
- Office of the Principal Scientific Adviser to the Government of India
- Main warning
- Strict localisation may raise compliance burdens for startups and MSMEs and invite trade retaliation.
- Proposed framework
- A four-tier risk-aligned matrix linking data sensitivity to transfer pathways.
- Economic estimates
- The paper cites modelling that balanced approaches could increase global GDP by about 1.77% and exports by about 3.6%, while fragmented regimes could reduce them by about 4.5% and 8.5%.
- India’s digital economy
- It accounted for about 11.74% of GDP, or approximately Rs 31.64 lakh crore, in 2022-23, according to the paper.
- Proposed institution
- A central interdisciplinary AI-data interoperability coordination body within the National Data Governance Committee.