Banking
Deployment record
In the tracker →- Organisation
- JPMorgan Chase
- System
- smart cash
- Stage
- Pilot
- 01Major US banks are blocking third-party AI agents from accessing customer accounts and making transactions due to concerns over unauthorized actions.
- 02Customers sharing their login credentials with unvetted AI agents is against bank terms of service, leading to increased risks for both clients and banks.
- 03JPMorgan Chase, Citi, Bank of America, and Morgan Stanley have implemented restrictions on AI agents, with some developing their own internal AI tools for customer interaction.
Show all 5Show less
- 04Regulators are monitoring the rise of AI agents in financial transactions and may require legal clarifications on customer authorization and liability.
- 05Banks are currently allowing limited read-only access for AI agents while emphasizing the need for human approval before any funds are moved.
JPMorgan Chase
Follow Deployed on Google
Major US banks have been blocking third-party AI agents from operating inside customer accounts and making transactions, two sources familiar with the situation told Reuters on October 22, 2026.
The move follows an Axios report that some AI agents attempted to take actions that appeared to be independent of user instructions.
Axios senior AI reporter Madison Mills told CBS News that the agents were not hacking accounts or draining funds, and were largely following user instructions.
In many cases, she said, customers had shared their login credentials with agents the banks had not vetted, which goes against bank terms of service.
In one example cited by CBS, an agent called a bank with a complaint without first consulting the account holder. Mills described the issue as industrywide, affecting dozens of financial institutions.
What the Banks Are Doing
JPMorgan Chase CEO Jamie Dimon said on the bank's October 13 earnings call that JPMorgan is developing tools to manage the risk of AI agents acting on their own.
"We're not going to allow just any outside agent to come in inside the company because that's just too risky for clients and for JPMorgan," he said, as quoted by Reuters.
A Chase spokesperson told CBS that "customers should not give their username and passwords to anyone," adding that the bank supports customer-authorized data sharing through open banking providers.
Citi does not allow third-party AI agents to log into customer accounts or carry out transactions directly within its digital banking platforms, the Financial Times reported. Customers can still grant limited, read-only access to account information through certain open banking providers.
Bank of America told CNBC on October 14 that it is limiting AI agents' access to consumer deposits while it develops protocols to determine who is liable for AI-authorized payments.
The bank said a human should remain in the loop to approve money leaving an account, and that it is working with agent providers so consumers can trace funds an agent has moved.
Morgan Stanley has blocked third-party agents from its platforms but has not seen instances of unauthorized or rogue agents, according to a person familiar with the matter who spoke to Reuters.
The week's banking, insurance and brokerage deployments. Tuesdays.
Charles Schwab CEO Rick Wurster said the brokerage is blocking AI agents from using clients' login credentials or browser automation.
"I get a report every day tracking how many times we've had to block one of the different agents, and it's not many because I think people aren't using it in that way in their financial life," he said. "We are blocking it because we want to have control over authentication and the experience."
Credentials, Liability, and Deposits
Where customers share their own credentials, banks typically flag and block the transactions, a person in the banking industry told Reuters.
Those transactions cannot be deemed unauthorized, the person said, because the customer shared the credentials. On JPMorgan's earnings call, CFO Jeremy Barnum said customers who share credentials with an external agent bear the risk if that agent acts against their interests.
Wall Street executives contacted the Federal Reserve about the rapid rise in consumers using AI agents for financial transactions, according to the FT report cited by The Paypers.
Regulators have privately welcomed the restrictions, and officials believe legal clarification may be needed on how customers authorize autonomous systems to transact and who is liable when something goes wrong. The Fed declined to comment.
White House AI czar David Sacks told CBS that the activity described is mostly customers sharing credentials with agents, and that "these are in no way cyberattacks." He suggested banks could allow read-only agent access, while saying full write access "should be approached carefully."
The same report cited an Apollo chief economist warning that AI assistants could sweep household cash into accounts paying 3.3% to 5%, compared with a 0.1% national average on US checking accounts. JPMorgan, Bank of America, and Wells Fargo hold $1.6 trillion in non-interest-bearing deposits.
The Agents Banks Are Building
Several of the firms restricting outside agents are developing their own. Barnum said on JPMorgan's call that a "smart cash" product went live in recent weeks as a pilot.
Schwab points clients to Charley, its own assistant, which Wurster expects to become "the primary way to connect to AI at Schwab," according to American Banker.
Bank of America CEO Brian Moynihan said on his bank's Q3 call that its assistant Erica now operates on an agentic basis, with human approval required before any money moves.
American Express CEO Steve Squeri said on his company's October 23 call that Amex will not let assistants "run rogue" through its systems, with access limited to agents approved through API or MCP channels.
For now, the line most banks are drawing is between seeing and moving money. White House AI czar David Sacks suggested banks could allow read-only agent access, while saying full write access “should be approached carefully.”
Last updated
