# Citi and Bank of America Restrict AI Agent Access to Client Accounts Major US banks block autonomous financial agents to prevent mass deposit migration and revenue loss. - **Source:** [INCRYPTED](https://brifing.si/source/7cd4625d-3f08-4cc1-983d-4c7d991839ba) - **Published:** 2026-10-18 - **Topic:** [Finance](https://brifing.si/c/finance) - **Original article:** - **On Brifing:** https://brifing.si/s/cfa3f343-dc8f-4af6-8eb4-551fa5e3efeb ## Restricting Autonomous Financial Access Citigroup and Bank of America have officially restricted third-party AI agents from accessing client accounts. These institutions are blocking autonomous assistants from executing independent transactions within digital banking services. This measure aims to protect user data and prevent unauthorized fund transfers initiated by AI systems without direct human oversight or explicit verification. ## Threat to Deposit Liquidity Banks fear that AI agents will automatically migrate household savings from low-yield checking accounts to higher-interest financial products. Currently, US banks hold approximately 1.6 trillion dollars in non-interest-bearing deposits. If these funds shift to accounts offering 3 percent yields, the sector could face an estimated 47 billion dollars in annual additional expenses. ## Regulatory and Operational Concerns In October 2026, banking sector representatives engaged with the Federal Reserve to address the risks posed by AI-driven financial operations. Regulators are currently evaluating the legal framework for autonomous system authorization. The primary concern remains establishing clear liability protocols for payments initiated by AI agents when errors occur or funds disappear unexpectedly. ## Human-in-the-Loop Requirement Bank of America CEO Brian Moynihan emphasized that human confirmation is mandatory for all fund transfers. The bank is actively collaborating with developers to establish strict rules regarding responsibility for AI-initiated payments. This strategy ensures that clients maintain full visibility and control over their assets, preventing confusion regarding where their money is moved. ## JPMorgan's Defensive Stance JPMorgan Chase CEO Jamie Dimon confirmed that the bank will not permit unrestricted access for third-party AI agents to internal systems. Given the potential for thousands of uncontrolled agents to operate simultaneously, the bank views such integration as an unacceptable risk. They are currently developing proprietary tools to protect customers from external AI interference. ## Yield Optimization Risks Economic analysis from Apollo suggests that AI agents could shift savings from accounts yielding 0.1 percent to those offering 3.3 to 5 percent. This automated optimization threatens the cheap deposit base that banks rely on for lending activities. Consequently, major institutions are prioritizing the defense of their existing liquidity pools against algorithmic financial management. ## Strategic Product Adjustments Christoph Stegmeier of Simon-Kucher suggests that banks must now actively incentivize clients to retain their savings. To counter the threat of AI-driven migration, institutions are expected to enhance loyalty programs and offer more competitive credit terms. Banks must make their deposit products inherently more attractive to prevent the mass exodus of capital. ## Future of Banking Integration The rise of AI assistants like Meta's Muse has accelerated the debate over wallet wars and automated payments. While banks currently favor read-only access for open banking services, the tension between innovation and risk management persists. The industry must balance the convenience of AI agents with the necessity of maintaining stable, low-cost deposit bases. ## Sources covering this story - https://incrypted.com/citi-y-bank-of-america-ohranychyly-dostup-yy-ahentov-k-schetam-klyentov/ — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._