🔍 Read the full analysis: AI Agents Can Help With Money—Here’s How To Set Them Up Safely 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 that sharing banking credentials with AI tools may leave them responsible for activity through that access. As firms including Visa and Robinhood discuss or offer agent-enabled financial services, safer approaches include limited permissions, spending caps and approval requirements—but those safeguards do not settle who bears losses.
Bank of America has warned customers not to share their banking usernames or passwords with AI agents, saying customers may be responsible for activity resulting from that access. The warning comes as financial companies weigh or offer ways for AI agents to handle payments and trades, putting a practical question in front of users: how to grant an agent useful access without handing it control of an entire account.
The email, titled “Protecting your finances in the AI World,” was posted by a customer on Bluesky. TechCrunch reported on it October 27, and a Bank of America spokesperson confirmed to the outlet that the email was authentic. The bank said credentials should never be shared with another person or an outside service, including an AI tool or agent acting on a customer’s behalf. It added that customers who share credentials are responsible for activity resulting from that access.
The bank also said it is “excited about what AI used responsibly can deliver.” Its warning does not set out a specific alternative access system for customers who want to use agents. Instead, it puts the emphasis on keeping passwords private and understanding the risks of allowing software to operate through a customer’s ordinary login.
Other financial developments underscore why the issue is pressing. The source reports that Visa told investors this week that trust is the biggest barrier to AI agents making payments. Robinhood said customers had placed more than $250 million into accounts that AI agents trade. That figure describes funds in those accounts; it does not establish how much was traded, the agents’ results, or the number of customers involved.
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 Full Account Access Raises the Stakes
A banking password is not a narrow permission. An agent logging in with a customer’s credentials may be able to carry out actions available to the customer, while the bank may have difficulty distinguishing the customer’s instructions from an agent’s activity—or from an intruder using stolen credentials. That can expose a user to both unwanted transactions and account locks if the bank flags the activity as suspicious.
Permission limits and liability are separate issues. A spending cap or a separate account may reduce the amount at risk, but it does not by itself determine whether a bank, technology provider or customer must cover a loss. Robinhood’s separate-account model includes spending limits, according to the source, but the company says customers “assume all risk” for trades made by AI agents. Users need to read the applicable terms rather than treating technical safeguards as a promise of reimbursement.
The concern also reaches beyond finance. Agents may process webpages, messages, invoices and documents to complete tasks. Malicious or misleading instructions embedded in those materials could steer an agent away from the user’s intent, a risk known as prompt injection. Giving an agent fewer capabilities can limit what such manipulation can accomplish, though it cannot guarantee safe behavior.
Popular right now
Roborock Q7 M5+ Robot Vacuum and Mop
Samsung 42-Inch Class OLED S90F 4K Smart TV (2025)
Dyson Cyclone V10 Origin Cordless Vacuum
As an affiliate, we earn on qualifying purchases.
Banks Confront Agent Logins
Axios reported October 22, citing unnamed sources, that major U.S. banks were finding third-party AI agents inside customer accounts, often using the customer’s own password. The report said banks may be unable to tell whether the activity was authorized by the customer. Those details are attributed to Axios’s sources; the source material does not identify the banks or provide public figures for how widespread the activity is.
The reported response has included blocking agents and revising terms. Bank of America’s customer email is a direct warning about credential sharing, but it does not establish a new legal ruling or a universal policy across banks. Separately, the OWASP project lists prompt injection among leading risks for AI applications, highlighting how an agent’s ability to read and act on outside information can create security problems.
Financial protections differ by jurisdiction and depend on the circumstances. In the United States, Regulation E excludes certain transfers initiated by someone given access by the consumer, while European Union payment rules require customers to take reasonable steps to protect security credentials. Whether and how those provisions apply to an AI agent has not been decided by a court or regulator, according to the source material.
““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 regulator or court has ruled, according to the source, on how U.S. Regulation E applies when a customer gives an AI agent banking access, or whether sharing credentials with an agent could amount to gross negligence under relevant EU rules. The legal outcome could depend on the facts, account terms and jurisdiction. The source author’s reading of those laws is an interpretation, not legal advice.
It is also unclear how many customers use agents to access bank accounts, how often agents have caused losses, and what protections will be offered across different banks and agent services. The reported $250 million in Robinhood accounts does not answer those questions. Nor does the source establish that every bank treats agent activity the same way.
Check Permissions Before Connecting
For users, the immediate step is to avoid giving an agent a bank password or other master credential. Where a provider offers them, narrower options can include read-only access, requiring approval for each payment, setting a capped virtual card, or using a separate account with only a limited balance. Some systems may support scoped tokens that work only with a particular merchant, amount and time window. Availability varies, and these controls should not be assumed to remove liability.
Before connecting any service, customers can review what it can see and do, the payment and trading limits, how access can be revoked, and who is responsible for unauthorized activity. Businesses building agent payments face the same design challenge: limit permissions to the task, preserve human approval for consequential actions, and make responsibility clear to users. Further bank policies, product safeguards and regulatory guidance may clarify the rules, but the source material gives no timeline for those developments.
Key Questions
What did Bank of America tell customers?
It warned customers not to share banking credentials with outside services, including AI-enabled tools or agents, and said customers may be responsible for activity resulting from access they share.
Is it illegal to give an AI agent a bank password?
The source does not establish that it is illegal. It says the application of U.S. and EU payment rules to AI agents has not been tested in court, and customers should review their bank’s terms and applicable rules.
What is a safer way to let an agent handle a payment?
Use the narrowest available permission, such as read-only access, approval for each payment, a capped virtual card, or a separate account with a limited balance. These measures can limit access but do not necessarily determine who covers losses.
What is prompt injection?
Prompt injection is an attempt to steer an AI agent through instructions hidden in material it reads, such as a webpage, email or document. Limiting an agent’s permissions can reduce the potential consequences, but does not eliminate the risk.
Does Robinhood cover losses from AI agent trades?
According to the source, Robinhood says customers “assume all risk” for trades by AI agents. The source does not describe how that term applies to every possible dispute or trading loss.
Source: ThorstenMeyerAI.com
