A Study on E-Commerce Chatbot Security, Memory Performance, and Complex Query Handling for Consumer Satisfaction

Abstract

AI chatbots are increasingly used in e-commerce customer service; however, issues related to data privacy, contextual memory, and complex query handling continue to affect user trust and experience. This study uses a survey of 160 digitally active Indian consumers to examine these challenges and their impact on consumer satisfaction. Analysis using chi-square tests of independence (α = 0.05) revealed statistically significant concerns: 76.3% of participants reported moderate to extreme anxiety related to phishing and data privacy; context loss affected 59.4% of respondents, with high frustration reported by 47.5%; and 76.9% preferred human assistance when chatbots failed to resolve complex matters. Overall consumer satisfaction remained only moderate (M = 3.49/5). The most desired enhancements were AI-based fraud detection and persistent memory solutions.

This study contributes by integrating security, memory performance, and complex query handling within a unified empirical framework in an emerging market context, offering actionable insights for improving chatbot reliability and user satisfaction.