FREE-AI Committee Report: RBI Directions On Implementation Of AI In Financial Services

Introduction

The Reserve Bank of India (RBI) on 13 August 2025 issued the report on the 'Framework for Responsible and Ethical Enablement of Artificial Intelligence' (FREE-AI Report) based on the recommendations proposed by the FREE-AI Committee (Committee). The Committee was constituted in December 2024 to identify risks stemming from Artificial Intelligence (AI) adoption and recommend regulatory frameworks on AI adoption in the financial sector.

Key findings from the RBI Survey on AI

The RBI conducted surveys covering regulated entities (including banks, NBFCs) and fintechs to understand the extent of AI adoption in the Indian financial sector and associated challenges. The key findings of the survey were that 20.8% of surveyed entities are deploying AI systems at present on customer support, sales, credit underwriting and cybersecurity. However, 67% of the surveyed entities expressed interest in exploring various AI use cases.

AI in finance: Opportunities & Challenges

The Report highlights following key opportunities and challenges in adoption of AI in financial sector:

Opportunities

Leveraging of AI for financial inclusion of underserved communities.
Integration of AI with Digital Public Infrastructure (DPI) for tailored and inclusive financial services.
Development of indigenous large-scale machine learning models.
Usage of AI agents for obtaining loan offers, undertaking comparative analysis and undertaking transactions real time.
Implementation of AI in emerging technologies such as quantum computing.

Challenges

Consumer risks and ethical concerns stemming from algorithmic bias.
Amplification of inaccuracies and algorithmic biases over high-volume transactions.
Excessive dependence on analogous AI models undermining market diversity.
Reinforcement of market trends, leading to market manipulation and volatility.
Difficulties in allocating liability among stakeholders.
Risk of non-adoption of AI on long-term competitive efficiency and inclusion goals.
Concerns for collusion amongst AI systems to maintain supra-competitive prices.
Cybersecurity vulnerabilities emanating from potential for misuse and cyberattacks.
Service interruptions and regulatory non-compliance by third party service providers.

Proposed amendment to existing laws

The Report acknowledges that the current legal framework, including the Information Technology Act, 2000 and rules thereunder, are sufficient to address current risks. The Report further analyses existing RBI guidelines and suggests the following amendments in respect of AI related aspects:

RBI Regulation	Amendments proposed
RBI Guidelines on Managing Risks and Code of Conduct in Outsourcing of Financial Services by Banks, 2006	Incorporation of AI-specific risks and AI-usage disclosure requirements.
RBI Cyber Security Framework in Banks, 2016	Inclusion of AI-specific threats (e.g., model poisoning, adversarial attacks) and incident protocols.
RBI (Digital Lending) Directions, 2025	Disclosure of AI-driven credit assessments; fairness audits to mitigate algorithmic biases.
RBI Master Circular on Customer Service in Banks, 2015	AI-usage disclosure requirements and establishment of processes for customers to contest AI driven decisions.
RBI (Fraud Risk Management in Commercial Banks (including Regional Rural Banks) and All India Financial Institutions) Directions, 2024	Implementation of AI-driven fraud detection, along with testing the accuracy and bias in these processes.
RBI (Information Technology Governance, Risk, Controls and Assurance Practices) Directions, 2023	Introduction of AI-specific access control measures for autonomous AI.
RBI (Outsourcing of Information Technology Services) Directions, 2023	Requirement of AI-usage disclosure by service providers and AI-specific risk assessments.

FREE-AI: The Seven Sutras

The Committee has laid down a set of the following seven overarching principles, "Sutras", to guide responsible AI innovation, governance and policy in the financial sector.

Public trust should be the foundation of AI systems.
Disclosure of usage of AI and allowing individuals the final authority to override AI systems.
Responsible, socially useful innovation should take priority over cautionary restraint.
AI systems should be designed and tested to promote fairness, equity and inclusion.
Entities deploying AI systems should be accountable for decisions of the AI systems, regardless of the level of autonomy of the systems.
AI systems should be understandable by design to entities deploying them.
AI systems should be safe, sustainable and resilient to physical and cyber risks.

Summary of Recommendations of the Committee

The Report contains several short-term and medium-term recommendations based on surveys and stakeholder discussions on aspects related to innovation and risk management. The firm was privileged to provide its recommendations to the Committee as a stakeholder.

Short-term recommendations

AI Disclosures: RBI-regulated entities (REs) to include AI-related disclosures in annual reports such as AI governance frameworks, AI adoption areas, consumer protection and grievance redressal measures.

Standing committee: Establishment of a standing committee to advise RBI on AI developments and regulatory needs.

Data Infrastructure: Building financial sector data infrastructure as a Digital Public Infrastructure (DPI) to support AI model development.

AI Inventories: Creation of AI inventories and sector-wide repositories covering models, use cases, dependencies, and risks.

Innovation Sandbox: Establishment of an AI sandbox for testing and developing AI solutions in a controlled environment

Medium-term recommendations

AI Liability Framework: Introduction of a graded liability and supervisory approach for AI.

Data and Cybersecurity Frameworks: Development of AI-related frameworks for data governance, cybersecurity, and consumer protection, with industry associations facilitating best-practice sharing.

Incident Reporting and Audits: Requirement of AI incident reporting and establishment of audit processes for REs.

Incentives for Indigenous AI: Promotion of developing of indigenous sector-specific AI models and ensuring equitable AI adoption.

Comprehensive RBI AI Policy: RBI to release an overarching AI policy balancing innovation with risk mitigation.

AI Compliance Toolkit: Development of toolkits to help REs align AI models and applications with regulatory expectations.

Board-approved AI Policies: Requirement of REs to frame board-approved AI policies covering governance, lifecycle management, risk controls, and third-party vendor liabilities.

Comments

The Report provides much-needed clarity on AI adoption in the financial sector. It acknowledges AI's transformative potential while setting out seven guiding Sutras and actionable recommendations to balance innovation with ethical principles of trust, fairness, and accountability. It also situates these sectoral principles within the broader national agenda, highlighting the IndiaAI Mission—the government's flagship programme to build a cohesive and strategic AI ecosystem. As a next step, financial sector stakeholders should work towards revisiting their AI service agreements and framing a board-approved policy which includes AI governance structures, creation of AI inventory and basic consumer protection measures.

The content of this document does not necessarily reflect the views / position of Khaitan & Co but remain solely those of the author(s). For any further queries or follow up, please contact Khaitan & Co at editors@khaitanco.com.