Banks like NatWest and Bank of America stated on Tuesday that using AI agents for online shopping might raise the risk of fraud, scams, and data-privacy breaches as they lay out guidelines for the technology’s development.
AI chatbots are being promoted as shopping tools by tech companies like OpenAI, Anthropic, Google, and Meta.
These companies foresee a future where customers utilize AI agents to choose products and make purchases on their behalf. Meanwhile, retailers are vying for control over the recommendations made by chatbots.
In September, British retailer John Lewis reported that queries coming from AI agents had increased to 2.5% from 0.3% a year prior, with the trend picking up speed.
Customers were excited about the potential of agentic commerce and eager to enable it, according to a report from the group of banks, which also includes ING, New Zealand’s ASB Bank, US lender Capital One, and Commonwealth Bank of Australia.
But they cautioned that technology was developing more quickly than consumer protections and business norms.
According to the survey, consumers are unsure if AI will operate in their best interests. They worry that AI agents can make poor purchases, overspend, or, worse, lose their money to fraud and scams. They don’t know who they should contact if something goes wrong or if they will be protected.”
Risks mentioned in the paper included AI agents asking for consumers’ card information and entering it directly into websites or directing users to less secure payment options.
The banks want to make several recommendations to legislators, including mandating disclosure when an AI agent participates in a transaction, increasing transparency about AI agents’ decision-making processes, and ensuring client data security.
The research stated that while various systems should be compatible, consumers and merchants should also be allowed to select which AI-powered e-commerce services they utilize.
