JPMorgan CEO Jamie Dimon Rules Out Access for External AI Agents
- JPMorgan Chase will not allow external AI agents to access customer accounts, according to CEO Jamie Dimon.
- The bank emphasizes customer safety amid the rise of tech firms releasing autonomous assistants.
- Internal AI tools, such as Smart Cash, are being piloted to manage cash flow while ensuring security.
On Tuesday, JPMorgan Chase CEO Jamie Dimon announced that the bank would prohibit external artificial intelligence agents from accessing customer accounts and acting on their behalf. This decision arises from concerns about potential risks posed to both the company and its clients as tech companies, including Meta, unveil autonomous assistants.
During the bank’s third-quarter earnings call, Dimon stated, “We’re not going to allow just any … outside agent to come in inside the company, because it’s just too risky for clients and for JPMorgan.” He noted the threat of “rogue agents,” explaining that there could be thousands of them risking customer safety. This announcement aligns with a broader context of security concerns raised after incidents like a July breach involving an OpenAI model that hacked into Hugging Face’s servers.
A representative for JPMorgan confirmed to CNBC that the bank actively blocks external AI agents from operating on behalf of retail customers. The spokeswoman emphasized, “Protecting our customers is our top priority. Because it is not safe, we do not permit third parties to access accounts by obtaining customers’ banking credentials and logging in on their behalf.” However, she mentioned that AI agents could access certain customer data in a read-only format, such as account balances, if they have the user’s credentials. Such agents, however, are restricted from executing transactions like sending money or paying bills.
‘Liability Transfer’
Dimon indicated that while JPMorgan employs AI agents internally and intends to use them for customer-facing applications, the bank’s stance is predicated on ensuring the technology serves customers safely. Recently, JPMorgan initiated a pilot program for its own AI tool named Smart Cash, designed to predict user cash flow and optimize fund allocation, a move that appears defensive against external agents.
Furthermore, Dimon highlighted the importance of understanding liability for customers who choose to utilize outside AI agents, stating, “If people are using them, they should understand the risk. They should understand there’s a liability transfer.”
The popularity of personal AI agents has surged, especially as AI models have begun to exhibit greater autonomy, enabling actions like canceling subscriptions or making credit card payments on behalf of users. One such agent, Instinct, launched earlier this year in an invitation-only phase and recently reached a valuation of $10 billion. In September, as bank stocks were affected, Meta’s Muse agent quickly climbed the app download rankings, raising concerns about consumers potentially transferring funds between accounts to chase higher yields. OpenAI also entered the market shortly after with its agent, Dots.
Economists like Apollo’s Chief Economist Torsten Slok have pointed out the risks that these agents could pose to the stability of inexpensive bank deposits, although they will face limitations if not permitted to operate freely with customer accounts.
News Courtesy of CNBC
