New Survey: U.S. Consumers Are Wary of AI Agents Handling Their Money, Bank Account Information

As a growing number of companies offer artificial intelligence (AI) agents to carry out everyday tasks, a new national survey conducted by Morning Consult on behalf of the American Bankers Association finds that Americans are wary of using them to help manage their personal finances. Key takeaways from the survey, unveiled today at ABA’s 2026 Annual Convention, include: 

• Most adults say consumers shouldn't have to give AI their bank information, and that banks should be able to require security protections from AI agents.  
  • Four-in-five adults (79%) agree consumers should not be required to share their online banking username and password with an AI tool to get help managing their finances, and they think a bank should be able to know whether it is dealing with a customer or the customer’s AI agent (80%)
  • Three-quarters (76%) also say banks should be able to require AI agents to show adequate privacy and security protections before giving them access to customers' personal information
  • Three-quarters (74%) say they care if the business that handles their finances is held to the same legal and regulatory requirements as a bank
• Most Americans are uneasy about AI handling their money. 
  • Three-quarters (77%) are concerned about giving an AI tool or agent permission to move money, make payments or carry out other financial transactions for them
  • Only 27% say they are likely to use an AI agent to autonomously manage cash balances in their bank account in the next 12 months, while 61% say they are unlikely
  • One-in-five (19%) say an AI agent has already asked them for login credentials or card numbers

“While agentic AI has the potential to be a compelling consumer tool, this survey shows that Americans are not ready to hand over their financial decisions to AI agents,” said ABA president and CEO Rob Nichols. “Consumers clearly still value the safety and security of their bank, and they are understandably wary of allowing AI to move their money or access sensitive account information. They also want their bank to protect them from the risks posed by AI.”

Americans Want Congress to Protect Local Lending in Establishing Digital Asset Rules

The survey also examined Americans’ attitudes towards regulating digital assets just weeks removed from the Senate debate over the Clarity Act. The research shows consumers want Congress to be careful in establishing rules for cryptocurrencies, stablecoin and other digital assets. and they strongly support protecting local lending and the financial system from risk. Key takeaways include:  

• By a 5-to-1 margin (66% vs. 14%), consumers agree that in establishing the first-ever rules for digital assets like cryptocurrencies and stablecoins in the U.S., Congress and the administration should be cautious and not take any action that could undermine our existing financial system, especially the community banks that help drive local economic activity in some parts of the country.
• 7 in 10 consumers (68%) say they would be concerned if banks had less funding available to make loans to individuals and businesses in their communities in the same manner they do today, a risk posed by language in the recently stalled Clarity Act.
• By a 4-to-1 margin (61% vs. 16%), consumers agree that Congress should prohibit crypto companies from being able to offer interest-like rewards for holding stablecoin if there is a risk it could draw away deposits from local banks and reduce the amount of funds available to banks to lend in the community and support economic growth.   

“As Congress considers the next steps for digital asset rules, consumers are sending a clear message that they don’t want legislation that could weaken community banks and undermine the economy,” said Nichols. “If Congress returns to the Clarity Act, this survey should remind lawmakers to avoid policies that could reduce the local lending that powers communities.”

The survey also asked about consumers' overall awareness and interest in digital assets. The responses suggest most Americans are not overly focused on crypto. The poll shows:

• 3 in 5 (59%) say digital assets are not relevant to their day-to-day financial life
• Consumer adoption of digital assets (like Bitcoin or stablecoin such as USDC) or non-fungible tokens (NFTs) remains low: 
  • Only 17% currently own digital assets
  • Only 30% have ever owned digital assets
  • 6 in 10 are unlikely to buy, hold or use digital assets in next 12 months

ABA released an accompanying infographic highlighting some of the survey results. The data released today are the latest in a series of results gauging U.S. consumers’ preferences and opinions regarding banks and their services. ABA recently released additional survey data showing that consumers recognize banks’ efforts to protect them from fraud and support policy actions to combat fraud. The full results for these survey questions are as follows:

When asked “How concerned would you be, if at all, about giving an artificial intelligence (AI) tool or agent permission to move money between accounts, make payments, or carry out other financial transactions on your behalf?” consumers provided the following answers:

• Very concerned – 52%
• Somewhat concerned – 25%
• Not too concerned – 10%
• Not concerned at all–4%
• Don’t know – 10%

When asked “How likely are you to use an artificial intelligence (AI) agent to autonomously manage cash balances in your bank account and perform financial tasks such as redirecting or distributing deposits to High-Yield Savings Accounts or Money Market Funds within the next 12 months?” consumers provided the following answers:

• Very likely – 11%
• Somewhat likely – 16%
• Somewhat unlikely – 14%
• Very unlikely – 47%
• Don't know/no opinion – 12%

When asked “Has an artificial intelligence (AI) agent ever asked you to provide login credentials for your financial accounts and/or your debit or credit card numbers?” consumers provided the following answers:

• Yes – 19%
• No – 62%
• Don’t know/No opinion – 18%

When asked “Do you agree or disagree with this following statement? Consumers should not be required to share their online banking username and password with an artificial intelligence (AI) tool, app or service in order to receive help managing their finances.” consumers provided the following answers:

• Strongly agree – 58%
• Somewhat agree – 21%
• Somewhat disagree – 6%
• Strongly disagree – 4%
• Don't know – 11%

When asked “Do you agree or disagree with this following statement? A bank should be able to know when it is dealing directly with a customer and when it is interacting with the customer’s artificial intelligence (AI) agent.” consumers provided the following answers:

• Strongly agree – 57%
• Somewhat agree – 23%
• Somewhat disagree – 7%
• Strongly disagree – 3%
• Don't know – 11%

When asked “Do you agree or disagree with this following statement? A bank should be able to require artificial intelligence (AI) agents to demonstrate adequate privacy and security protections before giving them access to personal information of customers.” consumers provided the following answers:

• Strongly agree – 55%
• Somewhat agree – 21%
• Somewhat disagree – 6%
• Strongly disagree – 5%
• Don't know – 13%

When asked “Which of the following comes closest to your view, even if neither is exactly right?” consumers provided the following answers:

• I care if the business that handles my finances is held to the same legal and regulatory requirements as a bank  – 74%
• I don’t care if the business that handles my finances is held to the same legal and regulatory requirements as a bank  – 8%
• Don't know/No opinion – 17%

When asked “Do you agree or disagree with this following statement? In establishing the first-ever rules for digital assets like crypto currencies and stablecoins in the U.S., Congress and the administration should be cautious and not take any action that could undermine our existing financial system, especially the community banks that help drive local economic activity in some parts of the country.” consumers provided the following answers:

• Strongly agree – 34%
• Somewhat agree – 32%
• Somewhat disagree – 9%
• Strongly disagree – 5%
• Don't know – 21%

When asked “Do you agree or disagree with this following statement? Congress should prohibit crypto companies from being able to offer interest-like rewards for holding stablecoin if there is a risk it could draw away deposits from local banks and reduce the amount of funds available to banks to lend in the community and support economic growth.” consumers provided the following answers:

• Strongly agree – 29%
• Somewhat agree – 32%
• Somewhat disagree – 11%
• Strongly disagree – 5%
• Don't know – 23%

When asked “How concerned would you be, if at all, if banks had less funding available to make loans to individuals and businesses in their communities in the same manner they do today?” consumers provided the following answers:

• Very concerned – 31%
• Somewhat concerned – 37%
• Not too concerned – 12%
• Not concerned at all–7%
• Don’t know – 13%

When asked “Do you own any digital assets such as cryptocurrencies (like Bitcoin or stablecoin such as USDC) or non-fungible tokens (NFTs)?” consumers provided the following answers:

• Yes, I currently own digital assets – 17%
• No, I have previously but do not currently own digital assets – 13%
• No, I have never owned digital assets – 59%
• Don’t know – 11%

When asked “How likely are you to buy, hold or use digital assets such as cryptocurrencies in the next 12 months?” consumers provided the following answers:

• Very likely – 13%
• Likely – 15%
• Unlikely – 16%
• Very Unlikely – 44%
• Don't know – 12%

When asked “How relevant are stablecoins, cryptocurrencies and other digital assets to your day-to-day financial life?” consumers provided the following answers:

• Very relevant – 12%
• Somewhat relevant – 15%
• Not too relevant – 14%
• Not at all relevant – 45%
• Don’t know – 14%

About the Survey
This poll was conducted by Morning Consult on behalf of the American Bankers Association from October 9-12, 2026, among a national sample of 4,402 adults split into two representative groups for specific question sets (Split Sample A n=2,191 | Split Sample B n=2,211). The interviews were conducted online and the data were weighted to approximate a target sample of adults based on age, race/ethnicity, gender, educational attainment, and region. Results from the full survey have a margin of error of plus or minus 1 percentage point.