🔍 Read the full analysis: AI And Your Bank Account: Share Less, Set Clear Limits on ThorstenMeyerAI.com
Turn your quiet moments into listening time
- Thousands of audiobooks, podcasts and originals
- Listen on your phone, tablet or Echo — also offline
- Cancel anytime
TL;DR
Bank of America has warned customers not to give banking credentials to AI assistants and says customers are responsible for activity resulting from shared access. The warning comes as banks report difficulty distinguishing customer-directed AI agents from unauthorized account access, while legal protections and liability for agent-initiated transactions remain unsettled.
Bank of America has warned customers not to share their usernames or passwords with AI assistants, saying customers are responsible for activity resulting from that access. The email, confirmed as authentic by the bank to TechCrunch, arrives as financial firms grapple with AI agents accessing accounts and making transactions, with questions about customer protections still unresolved.
The email, titled “Protecting your finances in the AI World,” says credentials should never be shared with another person or provided to an outside service, including an AI-enabled tool or agent acting on a customer’s behalf. It adds that customers who do share credentials are responsible for activity resulting from that access. Bank of America also said it is “excited about what AI used responsibly can deliver.”
The warning addresses a practical problem: an agent logging in with a customer’s password may appear to the bank to be the customer, or someone who has stolen the credentials. That can make it difficult to distinguish an authorized task from account misuse. A customer-shared password may also give an agent broader access than its intended job requires.
Other developments underline that AI agents are already interacting with financial services. The source report says Visa told investors that trust is the biggest barrier to agents paying for purchases, while Robinhood said customers had placed more than $250 million into accounts that AI agents trade. The report does not specify the time period for that Robinhood figure, so it should be treated as a reported total rather than a rate of growth.
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
Who Bears the Risk of Agent Payments
The dispute is not just about whether an agent can complete a task. It is also about who pays when it makes a mistake, performs an unwanted transaction or is manipulated into acting. When an agent uses a customer’s login, the bank may have limited ways to identify which actions the customer intended. A dispute could then turn on the circumstances of access and the terms governing the account.
Bank of America’s warning makes the customer-facing risk explicit, but it does not settle how consumer-protection laws apply to AI agents. The source report’s author reads US and EU rules as potentially relevant, while noting that no court or regulator has ruled on the specific AI-agent question, as far as the author could establish. That distinction matters: the bank’s email is a statement of its position, not a judicial determination of liability.
For customers, the practical choice is between handing over a broad credential and using narrower controls where available. For financial firms and AI developers, the challenge is to let agents carry out approved tasks without giving them unrestricted access or leaving customers unclear about responsibility.
Products Worth Considering
Fitbit Versa Smart Watch – Black Aluminum, One Size (S & L Bands Included)
BowFlex SelectTech 840 Adjustable Kettlebell
BowFlex Results Series SelectTech Dumbbells
As an affiliate, we earn on qualifying purchases.
Banks Confront Customer-Logged-In Agents
On October 22, Axios reported, citing unnamed sources, that major US banks were finding third-party AI agents inside customer accounts, generally logged in with the customer’s own password. The report said banks often could not tell whether the activity reflected the customer’s instructions. The October 27 report by TechCrunch on Bank of America’s email was based on a customer-posted screenshot; a bank spokesperson confirmed the email was authentic to TechCrunch.
The source material describes several ways to limit an agent’s authority rather than sharing a full banking password: read-only access, approval before each payment, capped virtual cards, and tokens limited by merchant, amount or time. A separate account or pool of funds can also limit potential losses. These controls can reduce the scope of access, but they do not by themselves answer who is legally responsible if something goes wrong.
The report also identifies prompt injection as a security concern: instructions hidden in a webpage, email or document may influence an AI agent that reads that material. The article cites OWASP’s AI application risk list, but the supplied material does not include further technical detail or a specific incident involving a bank customer.
““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 the customer email confirmed to TechCrunch
Liability Rules Have Not Been Tested
It remains unsettled whether sharing banking credentials with an AI agent changes a customer’s legal protections in a specific case. The source report says US Regulation E excludes certain transfers initiated by a person given account access by the consumer, but applying that language to an AI agent or its operator is the author’s interpretation. The report says no regulator or court has ruled on that question, as far as the author could find.
For the EU, the report cites Payment Services Directive provisions requiring customers to take reasonable steps to keep personalised security credentials safe, with potential liability in cases involving intent or gross negligence. Whether giving credentials to an AI agent meets that standard has not, according to the source, been tested in court. The legal outcome could depend on facts and applicable rules; the article is not legal advice.
The source material does not give details about the Robinhood accounts’ time period, the specific safeguards in use across all relevant products, or the scope of Bank of America’s email distribution. It also does not identify a confirmed case in which an AI agent caused a customer loss under these circumstances.
Watch for Tighter Payment Controls
Customers considering AI-powered financial tools can check whether a service requires their bank password, what transactions it can initiate, and how to revoke access. Where supported, limited permissions, transaction approvals, spending caps and separate funds restrict the agent’s reach more than a full account login. Customers should also review the provider’s terms to see how it describes responsibility for agent actions.
Financial firms are likely to face continued pressure to distinguish customer-authorized agents from suspicious access and to explain how disputes will be handled. The source material does not provide a timetable for new rules, bank policies or product changes. The next meaningful developments will be clearer disclosures from banks and AI providers, any formal regulatory guidance, and evidence from disputes or court cases showing how existing protections apply.
Key Questions
What did Bank of America tell customers?
It warned customers not to share their banking username or password with outside services, including AI-enabled tools and agents. The email says customers are responsible for activity resulting from shared credentials.
Does the warning mean customers automatically lose fraud protection?
No such legal conclusion is established in the source material. The email states the bank’s position, while the application of US and EU consumer-protection rules to AI agents remains untested, according to the report.
What can customers use instead of sharing a password?
Where a provider and bank support them, narrower options include read-only access, approval for each payment, a capped virtual card, or a token limited to a particular merchant, amount or time period. A separate account with limited funds can also restrict exposure.
What is prompt injection?
It is a technique in which instructions placed in content an AI agent reads—such as a webpage, email or file—may steer the agent’s behavior. The source report identifies it as a risk but does not describe a confirmed banking incident caused by it.
Are AI agents already being used with financial accounts?
The source report says Robinhood reported that customers had put more than $250 million into accounts traded by AI agents. It does not state the period covered by that figure, and it also reports that banks have encountered agents logging into customer accounts with customer-provided passwords.
Source: ThorstenMeyerAI.com
