Robust and High-Quality Data key to Advanced AI Deployment in Banking and Finance
Published: Feb 18, 2026
By TIOLCorplaws News Service
NEW DELHI, FEB 18, 2026: AS part of the second day of the India AI Impact Summit 2026, the session "AI in Governance: Revolutionising Government Efficiency" brought together global researchers and senior policymakers to examine how artificial intelligence can strengthen public service delivery at scale. The discussion focused on moving beyond pilots and promise toward measurable impact, emphasizing rigorous evaluation, responsible deployment and systems-level readiness across government.
The session opened with a research presentation by Dean Karlan on the use of machine learning to improve targeting of public service delivery in Togo. The study demonstrated measurable improvements in food security, mental health and socio-economic indicators when AI-supported targeting was applied. At the same time, it revealed important limitations: phone metadata alone failed to capture treatment effects, exposing model drift and the challenges of predicting short-term vulnerability.
The findings reinforced the need for rigorous experimentation, iterative testing and evidence-based AI procurement. Speakers emphasized that AI systems must be carefully evaluated before large-scale deployment, particularly when they are used to identify beneficiaries, allocate resources or inform policy decisions.
The discussion then moved to a panel on India's preparedness for AI implementation in government. While progress is visible in building compute capacity, breaking data silos and upskilling public sector employees, significant challenges remain, particularly around scalability, data heterogeneity and ethical clarity. It was noted that only a small proportion of AI implementers fully understand their ethical frameworks, highlighting a governance gap that must be addressed alongside technical capability.
Assessing the current landscape Mohammed Y. Safirulla, Director, IndiaAI Mission, Ministry of Electronics and Information Technology (MeitY) pointed to areas such as banking and financial systems, including tax analytics and expenditure tracking, where AI adoption has advanced due to the availability of high-quality, structured data.
In contrast, sectors such as education and other citizen-centric services face greater complexity due to data heterogeneity and decision-making ambiguity. The need for carefully designed experiments, third-party audits and strong validation protocols was highlighted as critical to enabling scale. He concluded by recalling an unsupervised learning that was carried out during COVID and how the availability of high quality data enabled pre-emptive action to be taken.
A key theme throughout the session was the importance of addressing immediate operational pain points first, using AI to solve defined, high-impact problems before attempting broader systemic transformation. Speakers agreed that measuring intermediate outputs, conducting rigorous pilots and institutionalizing third-party audits will be central to building public trust and ensuring AI deployments are both effective and equitable.
The session underscored a central takeaway that AI can significantly enhance government efficiency and service delivery, but only when grounded in robust data, scientific validation and responsible governance frameworks.
