[Deutsche Bank and Ant Financial Release White Paper Proposing "Agentic Payments in Wholesale Finance"] Jinse Finance reported on November 9th that Deutsche Bank and Ant Financial jointly released a white paper, "Agentic Payments in Wholesale Finance," discussing the application and governance framework of AI agents in institutional finance. The white paper outlines three scenarios: treasury and liquidity management, which can predict liquidity gaps and surpluses, evaluate solutions, and execute internal transfers, investments, financing, and foreign exchange operations; cross-border payment orchestration, which can balance exchange rates, channels, and execution timing to optimize costs, speed, compliance, and settlement risks; and trade finance, which can interpret complex documents, identify data discrepancies, and generate financing or payment instructions after approval. In terms of governance, the two institutions advocate "boundary agent execution" rather than complete autonomy, adopting the IMF's three-tier architecture: intent and orchestration, control and authorization, and settlement. This separates AI reasoning from transaction execution and requires clear, limited, and revocable authorization limits, binding agent identity with accountable entities, and "Know Your Agent" (KYA) verification. The white paper uses Ant Financial's intelligent agent platform, Anvita Flow, as a case study. Sabih Behzad, a senior executive at Deutsche Bank, stated that the value lies not only in automatically executing payments, but also in how decisions are made, coordinated, and controlled.