AI-driven shopping surges, raising new risks for merchants

A new survey has revealed that the use of artificial intelligence in the shopping journey has become widespread, while simultaneously highlighting emerging risks for merchants as a result of agentic commerce.

Research conducted by Riskified, comprising input from more than 5,000 consumers globally, indicates that 73% of shoppers now utilise AI at some stage of their shopping experience. The report detailed that 45% of participants use AI assistants such as ChatGPT to seek product ideas, 37% rely on them to summarise reviews, and 32% turn to them to compare prices. Despite this high adoption rate, only 13% reported completing a purchase following a recommendation by an AI assistant.

The survey also shows a significant openness among consumers to increased AI involvement in their purchases, with 70% expressing at least some comfort with an AI agent completing purchases on their behalf. Furthermore, with the upcoming holiday season, 58% stated they are likely to employ such tools for gift shopping, pointing to the prevalence of AI technology as a key driver during this period. This growing use could present new liabilities for merchants, who may bear unforeseen exposure in this sector.

Emerging risks

"AI shopping agents may make buying easier for consumers, but they also blur the lines of accountability for fraud and policy abuse," said Jeff Otto, Chief Marketing Officer at Riskified. "When an AI shopping agent makes a purchase, is the merchant still liable when a dispute occurs - even though the shopper never visited the merchant's site during checkout? That uncertainty creates new risks for everyone involved. A disputed charge could stem from a hijacked AI account or even a legitimate customer claiming their AI assistant made a mistake. In each case, the liability remains murky, and data transparency is in short supply."

Riskified's analysis suggests that those conducting fraud, including sophisticated fraud rings, are positioned to take advantage of gaps made apparent by developments in agentic commerce. Already employing advanced AI techniques, these groups are expected to exploit vulnerabilities in identity verification, account security, and payment processing. For consumers, a smoother shopping experience may conceal an increase in disputes and chargebacks, shifting costs back to the merchant.

Consumer concerns

According to the survey, 32% of respondents cited payment security as their primary concern regarding AI-assisted shopping. This was followed by privacy worries at 26%, the potential for mistakes at 18%, and a perceived loss of control at 17%. Despite these concerns, consumer trust in AI as a tool for online purchasing is approaching that of traditional sales channels, with 36% stating they trust AI to influence purchases compared to 38% who rely on in-store advice. Only one in four prefer to shop online without any AI assistance.

Strategic actions for merchants

As consumer behaviour evolves with the adoption of the Agentic Commerce Protocol, an open standard facilitating purchases through AI interfaces, Riskified's recommendations focus on merchant preparedness. Fraud prevention teams are encouraged to proactively educate leadership on the risks and rewards associated with agentic commerce. Otto highlighted the need for strategic measures in facing these challenges:

"Fraud teams are the natural leaders for agentic commerce safety because they're hardwired to see both the opportunity and the risk. Their expertise isn't about saying 'no' to innovation; it's about building the guardrails that allow the company to say 'yes' confidently. They are essential to ensuring that exciting new channels like this become sustainable revenue drivers, not vectors for fraud and abuse."

Another key recommendation is for merchants to advocate for increased data transparency from technology platforms, referencing customer behavioural data such as device information and IP addresses as crucial to effective fraud management. Additionally, the analysis suggests greater collaboration with networked fraud intelligence platforms will be vital for merchants seeking to compensate for reduced direct data on consumer activity during agentic transactions.

Riskified's findings underscore a critical stage for eCommerce, as the rapid adoption of AI tools creates new opportunities along with new forms of risk for merchants, particularly regarding fraud liability and evolving consumer expectations around trust and transparency.