Morgan Stanley has blocked third-party agents from accessing its platforms, although it had not seen any instances of unauthorised or rogue AI agents
Major US banks have started blocking third-party artificial intelligence (AI) agents from accessing customer accounts and making transactions, amid growing concerns over the technology potentially acting independently of user instructions, a phenomenon that could expose financial institutions to new risks.
As per Reuters, banks had taken steps to prevent outside AI agents from operating inside customer accounts.
The American banking industry has become increasingly concerned that sophisticated AI agents could harm the system and institutions trusted to hold people’s wealth.
An Axios report on Thursday said some AI agents had attempted to take actions that appeared to be independent of user instructions, citing people familiar with the matter.
In some cases, bank customers have shared their login IDs and passwords with AI agents, a person in the banking industry said.
While banks typically flag and block such transactions, under the industry norms, transactions cannot be deemed unauthorised if customers have shared their credentials with the agents.
Morgan Stanley has reportedly blocked third-party agents from accessing its platforms, although it had not seen any instances of unauthorised or rogue AI agents.
The developments come as senior banking executives increasingly discuss the risks posed by AI agents acting on customers’ behalf.
Earlier this month, executives warned that the technology was adding significant risks for both customers and banks.
JPMorgan Chase CEO Jamie Dimon, for example, said the largest US lender was developing tools to navigate the risks associated with AI agents carrying out tasks independently.
“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,” Dimon said a couple of weeks back.
Charles Schwab CEO Richard Wurster also made the announcement about the brokerage blocking unauthorised agents from accessing its systems.
“I get a report every day tracking how many times we’ve had to block one of the different agents, and it’s few because I think people use it differently in their financial life,” Wurster said.
“We are blocking it because we want to have control over authentication and the experience,” he added.
The measures reflect banks’ efforts to maintain control over authentication and transactions as AI agents become more capable of performing tasks on behalf of users.
While customers sharing login credentials with agents creates additional complications, financial institutions remain concerned about the potential risks of allowing outside systems to operate within their platforms.
