Deutsche Bank and Ant Digital Technologies are examining how artificial intelligence could transform institutional payments, while the Federal Reserve warns about emerging risks. The next challenge is no longer whether AI can initiate financial transactions, but who controls its authority, verifies its decisions and accepts responsibility when something goes wrong.
Published: Tuesday, 17 November 2026
Category: AI & Blockchain
By: Akinyele Oluwale & Co. Investment Ltd.
Editorial focus: Agentic AI | Blockchain | Institutional Banking | Digital Payments | Financial Stability
Artificial intelligence is entering a new phase of financial development. Until recently, most discussions about AI in banking focused on customer service, fraud detection, credit assessment and financial forecasting. The conversation is now changing. Financial institutions are exploring systems in which AI agents can interpret instructions, evaluate alternatives and initiate financial activities within defined limits. This emerging field is known as agentic finance.
On 9 November 2026, Deutsche Bank and Ant Digital Technologies published a joint white paper titled Agentic Payments in Wholesale Finance. The paper examines how AI agents could support institutional treasury operations, cross-border payments and trade finance.
Our central assessment is that the next phase of AI and blockchain adoption will be determined by governance, financial controls and institutional accountability not automation alone.
Consider a multinational company operating across Nigeria, Europe and Asia. Its treasury department must manage cash balances, pay suppliers, monitor exchange rates and ensure that sufficient liquidity is available to meet financial obligations. Traditionally, these activities involve finance professionals, banking platforms and established approval procedures. An AI-powered system could potentially analyse cash positions, identify payment obligations and recommend efficient funding arrangements. A more advanced AI agent might also prepare payment instructions for execution.
The distinction is significant. A system that recommends a transaction creates one category of risk. A system authorised to initiate transactions creates additional financial, legal and operational responsibilities. For example, an AI agent might correctly identify an outstanding supplier invoice but what happens if the invoice is fraudulent? What if the supplier's banking information has been changed without authorisation? What if the transaction exceeds an approved spending limit? These questions demonstrate why intelligent automation must operate within enforceable controls.
The challenge is not simply developing an AI model capable of processing financial instructions.
The challenge is ensuring that every financial action remains authorised, traceable and accountable.
The November 9 white paper represents an important contribution to institutional discussions about AI-powered financial operations. It examines potential applications across three major areas.
Treasury management: AI agents could help institutions monitor liquidity requirements, evaluate funding options and coordinate financial activities. Cross-border payments: AI could assist with payment routing, transaction preparation and the management of operational requirements across jurisdictions. Trade finance: AI agents could support document processing, transaction monitoring and coordination between businesses and financial institutions. However, the white paper does not establish that fully autonomous institutional payments are already operating at scale. It presents potential applications and governance principles that require further development and evaluation. One particularly important consideration is the separation between an AI agent's recommendation and the institution's authority to approve a transaction. An agent may identify an opportunity or prepare a payment instruction but independent controls should determine whether the instruction is permitted. This distinction protects financial institutions from treating AI-generated decisions as automatically valid. Institutional implication: Banks are beginning to examine how AI can support financial execution without surrendering control over financial authority.
On 16 November, Federal Reserve Governor Michael Barr highlighted risks associated with the increasing use of artificial intelligence across the financial system. His remarks followed the Federal Reserve's latest Financial Stability Report, which identified growing concern about AI-related vulnerabilities. Among surveyed financial-system contacts, 87% identified AI as a significant emerging risk, compared with 50% in May. This reflects respondents' concerns, not a measured probability of financial losses. Barr raised several important issues. AI systems could introduce new cybersecurity vulnerabilities. Automated trading strategies could contribute to correlated market behaviour. Financial institutions may struggle to establish whether an AI agent accessing a customer's account has the necessary authority and increasingly automated financial activity could create challenges for existing safeguards.
These concerns are particularly relevant when AI systems move beyond generating recommendations and begin initiating financial transactions. The warning does not mean AI should be excluded from banking. It means financial institutions must establish safeguards proportionate to the responsibilities delegated to automated systems. Institutional implication: AI adoption is becoming a financial stability and governance issue, not merely a technology investment decision.
Another development emerged on 10 November, when Reuters reported findings from McKinsey's Global Banking Annual Review concerning the rising cost of AI agents. The analysis suggested that spending associated with AI systems could become a significant component of banking operating costs. Some institutions are therefore developing more structured approaches to measuring AI expenses and evaluating the economic returns from automation.
Blockchain could become useful in selected AI-driven financial arrangements, particularly where multiple parties require shared transaction records or programmable digital settlement. For example, a properly authorised AI agent might initiate payment for a digital service using an approved stablecoin arrangement. A blockchain network could record the transfer. Smart contracts could enforce certain predefined transaction conditions. However, these capabilities do not solve every problem. A blockchain can record that a transaction occurred. It cannot independently establish whether the underlying commercial instruction was legitimate. A smart contract can execute predefined rules. It cannot guarantee that the information supplied to those rules is accurate and a stablecoin may provide a digital settlement mechanism without eliminating issuer, redemption or regulatory risks. This creates an important distinction between transaction execution and transaction legitimacy.
The strongest financial systems must address both.
Layer | Essential requirement |
|---|---|
Identity | Establish which person or institution the AI agent represents |
Authority | Define what the agent is permitted to do |
Verification | Confirm transaction details and applicable conditions |
Settlement | Transfer value through an appropriate payment arrangement |
Accountability | Maintain records, oversight and procedures for resolving errors |
Blockchain may support the settlement and recordkeeping layers but identity, authorisation and accountability require additional institutional arrangements.
AI can automate financial decisions. Blockchain can support digital settlement. Neither technology removes the need for responsible financial governance.
Nigeria's financial technology industry could eventually benefit from advances in AI-powered financial services. Potential applications include invoice processing, business payment reconciliation, customer support and financial administration. For small and medium-sized businesses, improved automation could reduce some operational burdens. However, Nigeria's financial environment presents important practical considerations. Businesses must manage payment reliability, fraud risks, customer identification, data protection and regulatory compliance. Introducing AI agents into these processes creates additional responsibilities. For example, an automated system that processes supplier payments must operate within approved financial limits. It should not independently change beneficiary details or approve unusual transactions without appropriate controls. Similarly, a blockchain-based payment system must comply with applicable Nigerian laws and financial regulations.
Nigeria's November 2026 regulatory developments concerning stablecoins reinforce the importance of treating digital payments as regulated financial activities rather than purely technological experiments.
Our assessment is that Nigeria's greatest opportunity lies in developing responsible AI-enabled financial infrastructure that improves efficiency while preserving trust and accountability.
At Akinyele Oluwale & Co. Investment Ltd., we believe the emerging relationship between artificial intelligence and blockchain deserves serious institutional attention. However, the industry must avoid confusing automation with financial progress. An AI agent that can initiate payments is not necessarily an improvement if the institution cannot explain, supervise or reverse an erroneous action where appropriate. A blockchain transaction is not automatically trustworthy simply because it is recorded permanently and a financial technology product does not become commercially valuable merely because it combines AI and blockchain. We believe five principles should guide responsible adoption.
First, authority must be explicit. AI agents should operate within clearly defined permissions. Second, financial controls must remain independent. Important transactions should be subject to appropriate approval and verification mechanisms. Third, accountability must be preserved. Institutions should remain responsible for the financial activities conducted through their systems. Fourth, technology must deliver measurable value. Automation should improve efficiency, reliability or customer outcomes. Fifth, innovation must remain consistent with regulation. AI-driven financial services should comply with applicable financial and data-protection obligations. The Deutsche Bank–Ant Digital white paper illustrates how institutions are beginning to examine these requirements. The Federal Reserve's warning demonstrates why the questions are becoming increasingly urgent.
Our institutional conclusion is that the future of financial AI will depend less on how independently machines can act and more on how reliably institutions can control their actions.
Several developments deserve attention during the remainder of November 2026 and beyond.
Institutional AI pilots: Watch whether banks move from governance frameworks toward controlled commercial deployments.
AI agent identity: Monitor standards for verifying which person or institution an agent represents.
Payment authorisation: Examine how financial institutions establish spending limits and approval requirements.
Blockchain settlement: Watch for credible examples of programmable settlement improving financial operations.
Stablecoin integration: Assess whether automated payments can reliably connect digital tokens with usable money.
Regulatory guidance: Monitor how financial supervisors address AI-driven financial activity.
Operational costs: Look for evidence that AI deployment delivers measurable economic benefits.
Cybersecurity: Assess whether institutions can manage the new risks introduced by automated systems.
The strongest evidence of progress will come from operational results, not promotional announcements.
Deutsche Bank and Ant Digital Technologies published an institutional agentic-payments white paper on 9 November 2026.
The paper examines potential applications in treasury management, cross-border payments and trade finance.
Federal Reserve Governor Michael Barr warned on 16 November about AI-related financial stability and transaction-authorisation risks.
Banks are also evaluating the rising costs of deploying AI agents.
Blockchain may support programmable settlement and transaction records in suitable financial applications.
Blockchain confirmation does not independently establish that a payment was authorised.
Institutional AI requires identity verification, financial controls, accountability and regulatory compliance.
Commercial success depends on measurable efficiency and reliable financial outcomes.
Akinyele Oluwale & Co. Investment Ltd. is an institutional financial intelligence platform covering global macroeconomics, central banking, institutional cryptocurrency, stablecoins, digital payments, tokenization, real-world assets, artificial intelligence and blockchain technology. Our mission is to provide evidence-based analysis that explains how financial institutions, emerging technologies and regulatory developments are reshaping global finance. We believe technological innovation should be evaluated through its economic usefulness, operational reliability and contribution to responsible financial development.
The latest developments in AI and blockchain reveal an important transition. Artificial intelligence is moving beyond analysing financial information toward supporting financial decisions and potentially initiating transactions. Blockchain technology may provide useful infrastructure for selected digital settlement arrangements but financial institutions cannot delegate accountability simply because a system is automated. Banks must know who authorised a transaction. Businesses must understand the financial obligations created by automated instructions. Regulators must ensure that appropriate safeguards remain effective and customers must have confidence that technology operates within clearly defined limits.
The Deutsche Bank–Ant Digital initiative and the Federal Reserve's latest warnings illustrate both sides of this transformation. The opportunity is substantial, but so is the responsibility.
The next financial revolution will not be defined by machines that can move money without human intervention. It will be defined by intelligent systems that can operate efficiently while remaining secure, authorised, transparent and accountable.
Observe the change. Understand the implications. Position with discipline.
Akinyele Oluwale & Co. Investment Ltd.
Where Global Finance Meets Tomorrow's Technology.
Website: www.akinyeleoluwale.finance
Editorial disclaimer: This publication provides general financial education and institutional analysis. It does not constitute personalised investment advice or a recommendation to purchase any financial asset.