Technology
Dimon Warns Banks Against Allowing AI Agents to Make Direct Payments
JPMorgan Chase CEO Jamie Dimon cautioned that allowing autonomous AI agents to execute payments would require handing over sensitive banking codes, effectively giving unvetted algorithms deep access to personal accounts without a guarantee of acting in the customer’s best interests. Speaking at The Economic Club of Chicago, Dimon said that such agents could perform actions that are "treacherous" for financial institutions. He warned that autonomous tools could enable round‑the‑clock cyber intrusions or trigger vulnerabilities if granted access to consumer accounts. Dimon’s remarks come as consumer‑facing AI agents that manage calendars, draft emails and book travel are being rapidly adopted. The JPMorgan chief had previously warned during the bank’s third‑quarter earnings call that the financial system could soon face thousands of rogue agents. Despite the warnings, JPMorgan is moving forward with its own controlled AI initiatives. CFO Jeremy Barnum confirmed that the bank’s proprietary Smart Cash tool, designed to help clients oversee liquidity across multiple accounts, is in an early‑stage pilot. JPMorgan has used AI internally since 2012 for risk assessment, fraud prevention, marketing, credit underwriting and customer support. Dimon compared the rapid rise of AI to the Industrial Revolution, describing the technology’s spread as a tsunami that cannot be stopped and urging public institutions and corporate leaders to manage the fallout. He noted that global governments cannot simply halt model proliferation, citing massive capital flows from tech giants such as Google, Amazon, Microsoft and Meta, who plan more than $700 billion in AI infrastructure spending this year. Recently, JPMorgan Chase, Citi Bank and Bank of America have banned AI agents from logging into customer accounts and moving money, fearing uncontrolled cash movements could erode vital deposit bases.
Frequently asked questions
Why did Jamie Dimon warn banks about AI agents making direct payments?
He said allowing autonomous AI agents to execute payments would hand over sensitive banking codes, giving unvetted algorithms deep access to personal accounts without a guarantee of acting in the customer’s best interests, potentially enabling round‑the‑clock cyber intrusions or triggering vulnerabilities.
What AI initiatives is JPMorgan Chase pursuing?
JPMorgan is moving forward with controlled AI initiatives, including an early‑stage pilot of its proprietary Smart Cash tool to help clients oversee liquidity across multiple accounts, and has used AI internally since 2012 for risk assessment, fraud prevention, marketing, credit underwriting, and customer support.
Which banks have banned AI agents from logging into customer accounts?
JPMorgan Chase, Citi Bank, and Bank of America have banned AI agents from logging into customer accounts and moving money.