RBI FREE-AI Framework

The Reserve Bank of India introduced the FREE-AI Framework - a set of 7 guiding principles ('Sutras') for responsible adoption of artificial intelligence in financial services, released in the August 2025 RBI report. This framework applies to all RBI-regulated entities deploying AI/ML models for functions including credit scoring, fraud detection, customer service automation, and risk assessment. The framework also explicitly mandates a graded liability system and the appointment of a Chief AI Ethics Officer (CAIEO) or equivalent for high-risk deployments.

1. Trust is the Foundation: Transparency and confidence in AI systems 2. People First: Human-in-the-loop and human judgment in AI deployment 3. Innovation over Restraint: Responsible enablement of AI innovation 4. Fairness and Equity: Bias mitigation across AI outcomes 5. Accountability: Clear ownership for AI outcomes 6. Understandable by Design: Explainability and interpretability of AI decisions 7. Safety, Resilience, and Sustainability: Cybersecurity, operational resilience, and energy efficiency

The FREE-AI Framework intersects with the DPDPA in several ways: (a) Data Governance aligns with DPDPA's consent and purpose limitation requirements; (b) Explainability relates to the Data Principal's right to information about processing; (c) for SDFs designated under Rule 13, algorithmic software monitoring obligations apply to AI/ML systems. Financial institutions using AI for decision-making about individuals must ensure that both frameworks are satisfied - the FREE-AI principles for the model itself and the DPDPA for the personal data it processes.

Financial institutions should: (a) conduct AI impact assessments covering both FREE-AI and DPDPA requirements; (b) document the lawful basis for personal data used in AI training and inference; (c) implement explainability mechanisms that also satisfy Data Principal access rights; (d) establish monitoring for bias and discrimination; and (e) maintain audit trails of AI-driven decisions.