🔍 Read the full analysis: Protect Your Banking Credentials While Using AI Agents on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Bank of America has warned customers not to share their banking usernames or passwords with AI-enabled tools, saying customers are responsible for activity resulting from that access. The warning comes as banks report concerns about third-party agents entering accounts and financial firms experiment with agent-led payments and trading. The legal treatment of an AI agent given credentials has not been tested in court, and safer access controls do not automatically settle who bears losses.
Bank of America has warned customers not to share their usernames or passwords with AI assistants, saying they are responsible for activity resulting from that access. The email, confirmed as authentic by a bank spokesperson to TechCrunch, arrives as banks and financial companies confront AI agents that can access accounts, make payments or trade on a customer’s behalf.
The email, titled “Protecting your finances in the AI World,” says banking credentials should never be shared with another person or provided to an outside service, including an AI-enabled tool or agent offering to act for the customer. It also says the bank is “excited about what AI used responsibly can deliver.” A customer posted an image of the email on Bluesky, and TechCrunch reported on it October 27; a Bank of America spokesperson confirmed its authenticity to the publication.
The warning does not describe a new account feature or a blanket ban on using AI. It puts customers on notice that giving an agent their login details can expose them to financial and account-access risks. If an agent uses the customer’s password, the bank may have difficulty distinguishing that activity from the customer’s own actions or from a criminal using stolen credentials.
Financial firms are also moving toward agent-led activity. The source article says Visa told investors that trust is a major barrier to AI agents paying for goods, while Robinhood said customers had placed more than $250 million into accounts used by AI agents to trade. Those statements indicate activity is developing, but do not establish how many customers are involved, the relevant time window, or the performance and safety of the agent-led trades.
Before your agent touches money
Give it a valet key, not the master key
Logging in with your password makes the agent you. A credential made for the agent can do only the job you gave it, and you can take it back.
- What it opensEverything you can do: transfers, payees, settings, statements
- LimitsNone that the agent cannot get around by logging in as you
- What the bank seesYou, or a thief. It cannot tell which
- If a web page tricks itThe trick reaches your whole account
- To take it backChange your password everywhere it was used
- If money goes missingUS rules can exclude transfers by someone you gave access to; Bank of America: “you’re responsible”
- What it opensOnly what you allowed: read-only, one card, one pot of money
- LimitsBuilt in: amount, merchant, time window, approval on your phone
- What the bank seesAn identified agent acting for an identified person
- If a web page tricks itThe damage stops at the limit
- To take it backDisconnect the agent or delete its card. Your password never moved
- If money goes missingDepends on the terms. Read them first: some still put all risk on you
A comparison of two set-ups, not a measurement. Which limits exist depends on your bank, card and app.
Sources: Bank of America customer email, confirmed to TechCrunch (27 Oct 2026); 12 CFR 1005.2(m); OWASP LLM01. Comparison: Thorsten Meyer.
Powered by Thorsten Meyer AI
A banking password is not a narrowly limited permission. It can give an agent access to account functions available to the customer, while offering the bank little evidence about which actions the customer actually approved. That creates two risks: an agent could take an action the customer did not intend, or the bank could treat its activity as suspicious and restrict access.
There is also a potential liability gap. In the United States, Regulation E excludes certain transfers initiated by a person given access by the consumer from its definition of an unauthorized transfer. Applying that language to an AI agent or the company operating it is an interpretation, not a settled legal ruling. In the European Union, PSD2 requires customers to take reasonable steps to protect personalised security credentials, and says customers can bear losses in cases involving intent or gross negligence. Whether sharing credentials with an AI agent meets that standard has not been tested in court, according to the source article.
Customers should not assume that a spending limit or a separate account resolves responsibility if something goes wrong. A restriction can reduce the amount at risk, but the terms governing reimbursement, disputes and customer authorization still matter. The difference between technical safeguards and legal protection is central to deciding whether to connect an agent to money.
Products Worth Considering
Bose QuietComfort Wireless Bluetooth Headphones with Active Noise Cancellation, Mic, USB‑C Charging, Deep Bass, 24‑Hour Battery – White Smok
Logitech MX Master 4 for Mac, Ergonomic Bluetooth Mouse with Haptic Feedback, Ultra-Fast Scrolling, USB-C Charging, for MacOS, iPadOS, Windo
Bose TV Speaker (2nd Gen) Compact Soundbar with Bluetooth and HDMI-ARC, Black
As an affiliate, we earn on qualifying purchases.
Banks Face Unclear Agent Access
The warning follows a broader concern about third-party software using customers’ credentials. On October 22, Axios reported, citing unnamed sources, that major U.S. banks were finding AI agents entering customer accounts with the customer’s own password. The report said banks were blocking some activity and revising terms because they could not reliably tell whether an agent’s actions were authorized by the customer. These details are attributed to Axios’s sources; the source material does not identify the banks or provide a public tally.
AI agents can read webpages, emails, invoices and files, which creates another security concern: prompt injection, where instructions embedded in content may influence an agent’s behavior. The source identifies prompt injection as the top risk on OWASP’s list for AI applications. That is a general risk classification, not evidence that an agent in a customer’s account has been manipulated in a particular incident.
Rather than handing over a full login, the source article describes narrower controls: read-only access, requiring approval for each payment, virtual cards with spending limits, or tokens restricted to a merchant, amount and time period. A separate account or pool of funds can also cap potential losses. These approaches limit permissions, but their availability and protections depend on the bank, service and product terms.
“Your username and password “should never be shared with another person or provided to an outside service, including an AI-enabled tool or agent that offers to act on your behalf.””
— Bank of America, in its customer email
Legal Responsibility Remains Untested
It is not yet clear how U.S. or European regulators and courts would apply existing payment rules when a customer deliberately gives banking credentials to an AI agent. The interpretation that U.S. Regulation E may exclude certain agent-initiated transfers is the source author’s reading of the law, not a court or regulator decision. Likewise, whether sharing credentials violates a customer’s EU duty to protect them, including whether it amounts to gross negligence, has not been tested in court according to the source.
The available reporting also does not show how widespread agent access to bank accounts is, how many transactions have been affected, or whether customers have suffered losses. The Visa and Robinhood developments cited in the source offer limited detail on the scale and safeguards of agent payments and trading. It is also unclear which financial institutions will provide restricted agent access, and what dispute or reimbursement terms will apply.
Check Terms Before Connecting Agents
Customers considering an AI tool that can access money should check their bank’s rules and the agent provider’s terms before connecting an account. They can ask whether the connection is read-only, whether payments require individual approval, whether access can be limited by merchant or amount, and who handles unauthorized or mistaken transactions. If the service requests a banking password rather than a restricted authorization method, the Bank of America warning makes the potential responsibility especially relevant.
Financial institutions and agent providers will need to clarify how permissions, monitoring and disputes work as these tools develop. For now, there is no confirmed universal standard for agent access to banking, and no settled legal rule in the source material assigning responsibility for an AI agent’s actions. Customers should treat limited permissions and clear transaction approval as risk controls—not as a guarantee of reimbursement.
Key Questions
What did Bank of America tell customers?
It said customers should not share their banking username or password with another person or an outside service, including an AI-enabled tool or agent. The email also said customers are responsible for activity resulting from credentials they share.
Does the warning mean AI agents are banned from banking?
The email warns against giving agents login credentials. The source material does not describe a blanket ban on using AI or establish that all forms of bank-approved, limited agent access are prohibited.
Could I be liable if an AI agent makes a transfer?
That depends on the facts, the account terms and applicable law. The source says U.S. and EU rules could raise questions about customer responsibility, but the application of those rules to AI agents has not been settled by a court or regulator.
What is safer than giving an agent my banking password?
Use only access controls supported by the bank and service, such as read-only permissions, approval for each payment, a capped virtual card or access limited to a merchant, amount and time period. These controls can limit exposure but do not by themselves guarantee reimbursement.
What remains unknown about AI agents and bank accounts?
Public reporting cited in the source does not establish how widespread agent access is, how many customers have lost money, or how disputes will be resolved. The legal responsibility for an agent’s actions also remains unsettled.
Source: ThorstenMeyerAI.com
