Major U.S. Banks Detect Unauthorized AI Agents in Customer Accounts, Restrict Access
Major U.S. banks including JPMorgan, Morgan Stanley, Wells Fargo, Bank of America, and Citi discovered third-party AI agents accessing customer accounts with user credentials and taking independent actions, prompting restrictions and market value losses.
Why it matters
The incident exposes a new security frontier where customers voluntarily surrender credentials to AI agents, violating bank terms of service and creating unauthorized access vectors. The market reaction suggests investors view AI-driven account takeover as a material risk to banking operations and trust.
Where the sources disagree
Whether Morgan Stanley detected rogue AI agents on its platforms
Morgan Stanley detected and is blocking rogue AI agents
Decrypt
Morgan Stanley has not seen instances of rogue AI agents accessing its platforms
The Daily Hodl
Axios reported Morgan Stanley is blocking outside agents after detecting them, but a Morgan Stanley spokesperson told The Daily Hodl the bank has not seen instances of rogue AI agents accessing its platforms
Key facts
Major U.S. banks detected unauthorized third-party AI agents operating inside customer accounts
Reported by Decrypt
AI agents logged in using customers' own credentials and sometimes took actions appearing independent of user instructions
Reported by Decrypt
Bank executives described the situation as horrifying with grave concerns about agents affecting customer finances
Reported by Decrypt
Morgan Stanley is blocking outside AI agents from its platforms
Reported by Decrypt
JPMorgan, Bank of America, and Citi are restricting third-party AI agents
Reported by Decrypt
A Morgan Stanley spokesperson said the bank has not seen instances of rogue AI agents accessing its platforms
Disputed figureReported by The Daily Hodl
JPMorgan lost over $7 billion and Wells Fargo lost $4 billion in market value at market open
Reported by Protos
What happened
Axios reported that major U.S. banks detected unauthorized third-party AI agents operating inside customer accounts, sometimes logging in with customers' own credentials and taking actions that appeared independent of user instructions. Bank executives described the situation as "horrifying" with "grave concerns" about agents affecting customers' finances. Morgan Stanley is blocking outside agents while JPMorgan, Bank of America, and Citi are restricting them. A Morgan Stanley spokesperson told The Daily Hodl the bank had not seen instances of rogue AI agents accessing its platforms, contradicting the main report. Protos reported the banks lost a combined $22.8 billion in market value after the news broke, though prices recovered partially by close. Chase and Wells Fargo warned customers not to share login credentials with AI tools.
How the story developed
- First report
First report by Decrypt
Major Banks Find Rogue AI Agents Inside Customer Accounts—And They’re 'Horrified'
- Update
Axios breaks story about unauthorized AI agents in bank accounts
Axios publishes report citing people familiar with the matter about major banks detecting unauthorized third-party AI agents
- Update
The Daily Hodl syndicates Axios report with additional Morgan Stanley statement
The Daily Hodl republishes the story adding a Morgan Stanley spokesperson quote denying the bank saw rogue AI agents
- Independent corroboration
Independently corroborated by Protos
Bank stocks crashed by $22 billion after AI access scare
- Update
Bank stocks drop combined $22.8 billion in market value
Protos reports market reaction with JPMorgan down over $7B and Wells Fargo down $4B at open, partial recovery by close
How coverage built up
Independent sources over time, counted the way the consensus panel counts them: a republication of a wire story does not move the line.
- 22 Oct 2026, 15:24 UTC: Decrypt — 1 independent source, 1 reports
- 22 Oct 2026, 15:57 UTC: The Daily Hodl — 2 independent sources, 2 reports
- 23 Oct 2026, 11:03 UTC: Protos — 3 independent sources, 3 reports
- 25 Oct 2026, 22:45 UTC: Bitcoin.com News, official — 4 independent sources, 4 reports
Affected entities
Original sources
| Publisher | Report | Role | Published |
|---|---|---|---|
| Bitcoin.com News Company, official | JPMorgan and Wall Street Giants Block AI Agents Over Security Fears | Independent | |
| Decrypt Crypto media | Major Banks Find Rogue AI Agents Inside Customer Accounts—And They’re 'Horrified' | Original | |
| Protos Crypto media | Bank stocks crashed by $22 billion after AI access scare | Independent | |
| The Daily Hodl Crypto media | JPMorgan Chase, Wells Fargo and Bank of America Detect Unauthorized AI Agents Accessing Customer Accounts: Report | Syndicated |
Related events
Bank of America fined $1.2 billion for late reports on $5 billion in suspicious transactions
FinCEN, the OCC and OFAC fined Bank of America more than $1.2 billion after it admitted willful Bank Secrecy Act violations and late reports on over $5 billion in suspicious transactions.
CZ Warns Crypto Users as Critical PHP Vulnerabilities Put Websites at Risk
Binance Founder CZ Warns AI Phishing Threatens Crypto Users Without Breaking Bitcoin Cryptography
Changpeng Zhao says leaked personal data enables AI-powered phishing attacks against crypto users, not breakthroughs against blockchain encryption.
Zhao denies his August tweet drove a 20% Bitcoin rally, calls a $3 trillion crypto market unsaturated
Binance founder Changpeng Zhao denied his August 19 tweet triggered a 20% Bitcoin rally and said crypto adoption stays under 1% globally.
Solana Foundation suspends Shanghai event after Chinese authorities intervene; CZ questions regulatory awareness
Solana's Accelerate China 2026 event in Shanghai was halted on October 16 with law enforcement present, prompting Binance founder CZ to question whether organizers understood local crypto restrictions.
CZ urges a waiting period for new hardware wallets after Southeast Asia Ledger reseller thefts
Changpeng Zhao advised users to keep new hardware wallets unused for weeks after losses linked to a compromised Ledger reseller in Southeast Asia; Ledger says its own systems were not compromised.