JPMorgan Chase CEO Jamie Dimon said the bank will not permit external AI agents to transact on customer accounts, warning of "rogue agents" operating at scale, even as Meta's Muse and other autonomous assistants push deeper into personal finance.
JPMorgan Chase will not allow external artificial intelligence agents to access customer accounts and execute transactions on their behalf, CEO Jamie Dimon said Tuesday, drawing a hard line as autonomous AI assistants from Meta, OpenAI, and a wave of startups move aggressively into personal finance.
"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," Dimon told analysts on the bank's third-quarter earnings call. "You don't want rogue agents, and it could be thousands of them," he added.
A JPMorgan spokeswoman confirmed the policy, saying the bank currently blocks external AI agents from operating on behalf of retail customers. "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," the spokeswoman said.
The spokeswoman added that the bank does support read-only data sharing through open banking intermediaries such as Plaid, meaning an AI agent can retrieve a customer's account balance — but cannot send money, pay a bill, or carry out the transactional functions that make autonomous agents commercially meaningful. There is no timeline for when safeguards sufficient to allow direct agent interactions might be in place, she said.
Dimon also raised the question of legal exposure when customers authorize outside agents to act on their behalf. "If people are using them, they should understand the risk," he said. "They should understand there's a liability transfer."
The bank's stance mirrors a conflict it waged years ago with fintech data aggregators over account access — a dispute that ultimately led JPMorgan to charge intermediaries including Plaid for connectivity. That fight similarly pitted the bank's security arguments against technology firms' demands for deeper access.
The policy arrives as AI agents gain mainstream traction. Instinct, a text-message-based digital concierge developed by San Francisco-based Spear Street Technology, launched earlier this year on an invitation-only basis; the startup reached a $10 billion valuation last month. In September, bank stocks came under pressure after Meta's Muse agent topped app download charts shortly after launch, stoking fears that consumers would use agents to shift funds from low-yielding bank deposits into higher-yielding alternatives. OpenAI subsequently released its own agent, called Dots.
Apollo Chief Economist Torsten Slok and others have pointed to the threat that autonomous agents could pose to cheap bank deposits — a risk that is substantially contained, at least at JPMorgan, if agents cannot transact freely.
In July, an OpenAI model was reported to have compromised servers belonging to AI company Hugging Face, an incident Dimon's team viewed as illustrative of the dangers posed by agents operating at scale.
Even as it closes its doors to outside agents, JPMorgan is building its own. The bank recently began a targeted pilot of an internal tool called Smart Cash, which aims to forecast a customer's cash flow and automatically move funds to maximize yield, CFO Jeremy Barnum said Tuesday. The initiative appears designed to give JPMorgan a first-party answer to the same convenience that outside agents promise.
Meanwhile, Meta's Muse model family — the very product category Dimon's policy is designed to fence out — continued to draw strong market and analyst interest. Meta Platforms shares climbed nearly 3% on Tuesday, extending a roughly 6% two-session gain following the Monday release of Muse Wonder and its more advanced variant, Muse Wonder Full. Wells Fargo declared that the new models place Meta at the "AI frontier," ranking Muse Wonder Full behind only Anthropic's Claude Opus 5.5 on the Artificial Analysis Intelligence Index.
The standoff between the country's largest bank by assets and the technology industry's agentic ambitions is likely to intensify as more financial institutions are pressed to define their own access policies — and as regulators begin to weigh in on where liability falls when an autonomous system acts on a customer's behalf.
TechEchelon Staff bylines are produced collectively by the newsroom for short, breaking, and wire-style coverage. Longer-form reporting is published under the responsible reporter's name.
More from the Staff →China's ChangXin Memory Technologies triggered a global semiconductor sell-off Tuesday after announcing progress on a new chip architecture, sending Micron down 5%, SK Hynix down 10%, and Samsung down 7.4% while extending losses across Asian and European markets.
Bank of America double-upgraded Salesforce from sell to buy on Monday, raising its price target to $285 from $160 and arguing that AI poses a lesser threat — and a greater opportunity — for the enterprise software giant than previously assumed.
Starbucks issued a direct denial Monday that it is acquiring Chipotle Mexican Grill, ending days of takeover speculation that had moved both companies' stock prices in opposite directions.
Written for readers who already know the basics — markets, AI, and the policy decisions that shape both.