AI agents for cash management in payment systems

Using prompt-based experiments with ChatGPT’s reasoning model, this paper evaluates whether a generative artificial intelligence (AI) agent can carry out high-level intraday liquidity management in a wholesale payment system. We simulate payment scenarios with liquidity shocks and competing priorities to test the agent’s ability to (i) maintain precautionary liquidity buffers; (ii) dynamically prioritize payments under tight constraints; and (iii) optimize the trade-off between settlement speed and liquidity usage. Our results show that even without domain-specific training, the AI agent closely replicates key prudential cash-management practices, issuing calibrated recommendations that preserve liquidity while minimizing delays. These findings point to the feasibility of automating routine cash-management tasks with general-purpose large language models, potentially reducing operational costs and improving intraday liquidity efficiency. We conclude with a discussion of the regulatory and policy safeguards that central banks and supervisors may need to consider in an era of AI-driven payment operations.