Fragmented government data threatens welfare delivery and AI adoption, says TBI report
Citing a NITI Aayog assessment linking poor data to 4–7% leakage in welfare spending, the report proposes seven measures to help states improve data quality, accountability and public-service delivery.
Citing a 2025 NITI Aayog assessment, the report estimates that poor-quality data causes fiscal leakage equivalent to 4–7% of India’s annual welfare spending.
NEW DELHI: Fragmented and poor-quality government data is weakening welfare delivery and could undermine India’s adoption of artificial intelligence in governance, according to a report released by the Tony Blair Institute for Global Change (TBI) on October 10.
The report, “From Digital Scale to Data Power: A Data Operating Model for India’s States”, argues that India’s extensive digital infrastructure has not been matched by a consistent ability to connect, verify and use administrative data across departments.
Citing a 2025 NITI Aayog assessment, the report estimates that poor-quality data causes fiscal leakage equivalent to 4–7% of India’s annual welfare spending. It identifies exclusion from welfare schemes, inefficient resource allocation and weaker public trust as other consequences.
The report cites previous beneficiary verification exercises to illustrate the financial implications. Removing 1.71 crore ineligible names from PM-KISAN saved an estimated ₹9,000 crore, while eliminating 3.5 crore bogus LPG connections saved ₹21,000 crore over two years. Removing 1.6 crore fake ration cards is saving approximately ₹10,000 crore annually, according to figures cited in the paper.
TBI said states generate substantial administrative data through beneficiary lists, health records, land registers, school enrolments and tax filings. However, these datasets are often collected for individual departments’ reporting needs rather than broader policy questions.
Vivek Agarwal, TBI country director and co-author, said governments needed to establish whether their data was reliable enough for either officials or AI systems to act on. Improving that foundation, he said, would support faster services, better decisions and earlier identification of problems.
The paper examines Union government initiatives led by the Ministry of Statistics and Programme Implementation, including NMDS 2.0, SQAF, the AI-Readiness Framework, the QPR Portal and the Model Data Sharing Framework.
It also draws on state initiatives such as Karnataka’s Kutumba social registry, Odisha’s Social Protection Delivery Platform and Rajasthan’s Pehchan Portal, alongside programmes in Andhra Pradesh, Uttar Pradesh, Tamil Nadu and Telangana.
The report proposes seven areas of action: clear leadership and mandates; data ownership and stewardship; common standards; shared data capabilities; independent quality assurance; legal and procurement safeguards; and a State Data Balance Sheet to track data assets and risks.
Co-author Ott Velsberg, Estonia’s former government chief data officer, said India’s next challenge was to make data consistently trustworthy, connected and useful for improving outcomes.
The report cautions that deploying AI on fragmented datasets could amplify existing gaps in government decision-making. It argues that stronger data governance is essential to translate India’s digital scale into more effective public services.
