970x125
To encourage the responsible and ethical adoption of Artificial Intelligence (AI) in the financial sector, a Reserve Bank of India (RBI)-constituted committee has recommended several measures, including the establishment of financial sector data infrastructure, data lifecycle governance, consumer protection and cyber security measures.
970x125
These proposals were put forward by an RBI committee, set up in December last year, to develop a Framework for Responsible and Ethical Enablement of Artificial Intelligence (FREE-AI) in the financial sector. The committee has given 26 suggestions which are based on six pillars – infrastructure, policy, capacity, governance, protection and assurance.
“A high-quality financial sector data infrastructure should be established, as a digital public infrastructure, to help build trustworthy AI models for the financial sector,” the committee said. This may be integrated with the AI Kosh – India Datasets Platform, established under the IndiaAI Mission.
The committee has recommended the establishment of an AI innovation sandbox, development of indigenous financial sector-specific AI models, adaptive and enabling policies and adoption of AI liability framework.
It has suggested that regulated entities should develop AI-related capacity and governance competencies for the board and C suite. Regulators and supervisors should also invest in training and institutional capacity building initiatives to ensure that they possess an adequate understanding of AI technologies and to ensure that the regulatory and supervisory frameworks match the evolving landscape of AI.
To ensure the safe and responsible adoption of AI within institutions, the committee has proposed that regulated entities should establish a board-approved AI policy which covers key areas such as governance structure, accountability, risk appetite, operational safeguards, and consumer protection.
Regulated entities should augment their existing business continuity plan (BCP) frameworks to include both traditional system failures as well as AI model-specific performance degradation.
Story continues below this ad
The committee has recommended that REs should implement a comprehensive, risk-based, calibrated AI audit framework, aligned with a board-approved AI risk categorisation, to ensure responsible adoption across the AI lifecycle, covering data inputs, model and algorithm, and the decision outputs.
970x125
