🔍 Read the full analysis: A Beginner’s Guide To Letting AI Agents Handle Financial Tasks on ThorstenMeyerAI.com
Get business pricing on office and shipping supplies
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Bank of America warned customers not to give banking credentials to AI tools, while Visa identified trust as a barrier to agent payments and Robinhood reported more than $250 million in accounts used by AI agents. Password sharing can expose users to account and liability risks; narrower permissions and human approval offer ways to limit access, but legal and consumer protections remain unsettled.
Bank of America warned customers not to share their usernames and passwords with AI assistants, as Visa and Robinhood signaled that AI agents are already entering financial activity. The bank said customers who share credentials are responsible for activity resulting from that access, while the legal treatment of an AI agent acting through a customer’s login remains unsettled.
The Bank of America email, titled “Protecting your finances in the AI World,” says login details should never be shared with another person or an outside service, including an AI-enabled tool or agent. A customer posted the email on Bluesky, TechCrunch reported it on October 27, and a Bank of America spokesperson confirmed to TechCrunch that it was authentic. The bank also said it is “excited about what AI used responsibly can deliver.”
The warning arrives as financial services companies address agent use from different angles. At an investor discussion on October 28, Visa described trust as the biggest barrier to AI agents making payments, according to the source material. Robinhood said customers had placed more than $250 million into accounts that AI agents trade. The material does not specify the period covered, how many customers or accounts are involved, or whether the figure represents deposits, assets, or another measure beyond money placed into those accounts.
The central security concern is that an agent using a customer’s password may appear to the bank as the customer—or as someone who stole the password. That can make it harder to distinguish approved actions from harmful ones and could lead to account restrictions. Separately, an agent can be influenced by instructions embedded in pages, emails or files it reads, a risk known as prompt injection. The source identifies it as a leading risk in OWASP’s list of AI application risks.
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
Why Agent Access Changes the Risk
Giving an AI agent a banking password can grant it the same broad access the customer has, rather than permission for one limited task. If the agent makes an unwanted payment or trade, the customer may face both the financial loss and a dispute about whether the activity was authorized. A bank may also respond to unusual activity by blocking access, even if the customer intended the agent to act.
Technical safeguards do not settle who bears the loss. Robinhood reportedly uses a separate account and spending limits for agent trading, yet the company says customers “assume all risk” for trades by AI agents. That statement illustrates why users need to read the service’s terms rather than treating limits as a promise of reimbursement. The amount of protection can depend on the product, the transaction and applicable law.
For consumers, the practical question is not simply whether an agent can perform a task. It is whether its access can be constrained, whether a person must approve consequential actions, and what happens if the agent or its provider makes a mistake. For companies building agent payments, trust depends on making those permissions and responsibilities understandable.
Popular right now
Shark Navigator Lift-Away Upright Vacuum (NV360)
Samsung 42-Inch S90H OLED 4K Smart TV Bundle
Roborock Q7 M5+ Robot Vacuum and Mop
As an affiliate, we earn on qualifying purchases.
The warning follows an October 22 Axios report, which cited unnamed sources, that major U.S. banks were finding third-party AI agents inside customer accounts. According to that report, agents often logged in using the customer’s own password, leaving banks unable to tell in some cases whether the activity reflected the customer’s intent. The source material says banks have been blocking some agents and revising terms in response.
U.S. Regulation E defines an unauthorized electronic fund transfer to exclude one initiated by a person given access to an account by the consumer, subject to conditions that include notifying the financial institution when that person is no longer authorized. The source’s author interprets this as potentially relevant to AI agents, but says no regulator or court has ruled on how the provision applies to an agent or the company operating it. In the EU, PSD2 requires customers to take reasonable steps to keep personalised security credentials safe. Whether giving credentials to an AI agent amounts to gross negligence under that directive has not been tested in court, according to the source.
These are legal provisions, not settled answers about every agent service or transaction. The source’s discussion is an interpretation, not legal advice, and readers’ circumstances may differ.
““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
Liability Rules Remain Untested
No court or regulator ruling is identified on whether a customer’s decision to share a password with an AI agent changes protections under the cited U.S. or EU rules. The legal outcome may depend on the agent’s role, the customer’s actions, the transaction and the terms of the financial service. The source’s interpretation should not be treated as a definitive legal conclusion.
Other details are also missing. The source does not provide the full terms governing Robinhood’s agent-trading accounts, the time period or accounting basis for the $250 million figure, or specifics about the agent activity banks reportedly detected. It is also unclear how widely banks are changing their policies and what protections other agent-payment products offer.
Safer Permissions and Next Steps
For now, the source recommends avoiding password sharing and looking for narrower ways to authorize an agent: read-only access, approval before each payment, a virtual card with a spending cap, or a token limited to a particular merchant, amount and time window. Another option is a separate account holding only money the user is willing to expose. These measures can limit potential damage, but they do not by themselves determine who is legally responsible.
Before connecting an agent, users should check what actions it can take, how to revoke access, what transaction limits apply and what the service’s terms say about losses. Financial providers and agent developers will need to clarify authorization and responsibility as they expand these tools. The source material does not identify a specific upcoming ruling, product launch or regulatory deadline; the immediate next step is for customers to review permissions and for providers to explain their safeguards and terms.
Key Questions
Should I give an AI agent my bank password?
Bank of America says customers should not share their credentials with outside services, including AI agents. A password can provide broad account access and make it harder for a bank to distinguish an agent’s actions from the customer’s.
Can a bank refuse to reimburse an AI agent’s transactions?
The answer is not settled for AI agents. The source discusses U.S. and EU rules that may affect responsibility, but says their application to agents has not been decided by a court or regulator. Check the financial provider’s terms and seek qualified advice about a specific dispute.
What is a safer way to let an agent handle a payment?
Consider limited permissions, such as read-only access, approval for each payment, or a virtual card with a cap. A token restricted by merchant, amount and time window can also reduce an agent’s access, if the provider offers it.
What does Robinhood’s $250 million figure mean?
The source says customers put more than $250 million into accounts that AI agents trade. It does not specify the reporting period, number of accounts or precise accounting basis, so the figure should not be read as a measure of profits or trading volume.
Source: ThorstenMeyerAI.com
