RBI banks on humans with AI safeguards for banking decision making framework

Industry experts have backed the Reserve Bank of India’s proposal to keep humans at the centre of AI-driven decision-making in banks, saying the draft Model Risk Management (MRM) framework provides the safeguards needed for wider adoption of artificial intelligence.

They said requirements such as human oversight, independent validation of AI models and ‘kill switches’ reinforce accountability while enabling banks to scale up using AI amid concerns over systemic risks from frontier models such as Anthropic’s Claude Mythos.