AI Usage Expands Among Merchants, Revealing Potential Risks
Artificial intelligence (AI) is revolutionizing the retail and e-commerce landscape, with financial companies rolling out platforms that allow AI agents to shop and make purchases with minimal customer interaction. These agentic AI systems are automating and optimizing transaction routing, fraud detection, and customer service.
Accelerating Transactions and Sales
Agentic AI is speeding up sales cycles by instantly generating quotes, suggesting complementary products, and facilitating complex ordering processes autonomously. This automation is particularly beneficial in B2B contexts, where orders can be complex and high-volume. Self-service portals empower buyers to access order history, modify orders, and manage returns efficiently, improving customer satisfaction and operational efficiency.
Fraud Detection and Prevention
The rise of agentic commerce introduces new fraud challenges as AI agents operate on behalf of users, making traditional fraud signals like device ID and user behavior less reliable. To combat this, AI-native, holistic fraud solutions are needed that continuously analyze risk throughout the customer journey. Systems like Ravelin use graph-powered machine learning to monitor user and agent actions in real time, adapt to new fraud patterns proactively, and maintain security without harming conversion rates.
Enhanced Customer Service
By 2029, a large majority of customer interactions—up to 80%—are expected to be handled by AI agents. This shift from reactive support to proactive management allows human agents to focus on complex cases while AI agents automate routine queries like order tracking and returns. Early adoption of agentic AI offers retailers valuable data insights into customer behavior and operational dynamics.
Future Implications
The future of retail and e-commerce will see a transformation as AI agents streamline transactions, provide personalized and quick customer service, and necessitate advanced, holistic fraud prevention systems customized for AI-driven interactions. Key implications include:
- From assisted to autonomous shopping: AI agents will shift buying from manual user-driven processes to fully autonomous transactions based on user preferences set once, moving customers into a hands-off shopping experience.
- Merchant competition shifts: Retailers will compete algorithm-to-algorithm, emphasizing data quality, trust signals, and machine-readable catalogs rather than traditional marketing messaging.
- Payment infrastructure evolution: Payment networks and platforms are developing specialized infrastructures to support real-time, AI agent-driven transactions securely.
- Enhanced fraud and risk management: Merchant strategies will need to balance seamless transaction flow with sophisticated, adaptive fraud detection that works in the agentic commerce environment.
AI in Practice
Prominent examples of AI in e-commerce include Walmart's Sparky and Amazon's Rufus, which are being used more prominently in the e-commerce apps of large retailers. However, it's essential to remember that AI should never be in the critical path of anything, as it is prone to making mistakes and "hallucinating" (making up things that aren't there). A buffer around any public-facing AI initiatives is necessary to prevent potential mistakes, overspending, or data breaches.
AI can also play a larger role in transaction routing, helping organizations select the most efficient payment method to cut costs and improve the customer experience. Machine learning in AI is based on experience with transactions that share similar attributes, but it struggles when encountering transactions never seen before. Nevertheless, AI's ability to handle broader amounts of data beyond the task at hand makes it useful for transaction routing, especially when cross-border elements come into play.
As the use of AI continues to be implemented in customer-facing situations, especially in areas like intelligent payments routing and agentic commerce, it's crucial to maintain a balance between automation and human oversight to ensure seamless, secure, and efficient transactions for both retailers and customers.
[1] "Current trends in agentic AI within retail and e-commerce." (Source) [2] "From assisted to autonomous shopping." (Source) [3] "AI in the Payments Ecosystem." (Report) [4] "Merchant competition shifts." (Source) [5] "Enhanced fraud and risk management." (Source)
Technology, specifically artificial-intelligence (AI), is set to play a significant role in payment infrastructure evolution, with systems developing specialized infrastructures to support real-time, AI agent-driven transactions securely. Moreover, in the realm of fraud and risk management, AI-native solutions are needed to combat fraud challenges introduced by AI agents, continuously adapting to new fraud patterns proactively and maintaining security while keeping conversion rates high.
