A group of major banks has warned that AI-powered shopping agents could introduce new risks for consumers, including scams, fraud, incorrect purchases and misuse of payment data, Reuters reported on September 22.
Banks including Bank of America, NatWest, ING, ASB Bank, Capital One and Commonwealth Bank of Australia contributed to a report examining the rise of so-called agentic commerce.
AI shopping agents are designed to search for products, compare options and in some cases complete transactions on a user's behalf. Banks said that convenience could create new questions about accountability when an AI agent makes an incorrect purchase, mishandles card details or directs users toward payment methods with weaker protections.
The report called for stronger industry and regulatory safeguards, including clear disclosure when AI is acting in a transaction, greater transparency in automated decision-making and stronger protection of customer data.
The banks also argued that consumers and merchants should retain freedom to choose among AI services and that payment systems should remain interoperable.
The warnings do not establish that AI shopping agents are inherently unsafe or that widespread fraud has already occurred. They identify risks that banks want policymakers and technology companies to address as agentic commerce grows.