AI and AR form powerful engagement loop in global e-commerce platforms

When users trust a platform, they are more willing to accept AI-generated recommendations and engage deeply with AR features. Trust reduces psychological resistance and perceived risk, allowing consumers to process system outputs more openly. In contrast, low trust weakens the effectiveness of even highly personalized or immersive systems.

Artificial intelligence (AI) and augmented reality (AR), when fully integrated, change not only how consumers shop but also keep them coming back, according to a new academic study published in the Journal of Theoretical and Applied Electronic Commerce Research.

The study titled “The Impact of Integrated AI and AR in E-Commerce: The Roles of Personalization, Immersion, and Trust in Influencing Continued Use” examines how AI-driven personalized recommendations and AR-based immersive experiences jointly influence long-term user engagement. The research analyzes survey data from 400 Chinese consumers with real AR shopping experience on Taobao, one of the world’s most advanced digital retail ecosystems.

The findings challenge several assumptions that dominate technology-driven retail strategy. While immersive experiences and visual novelty have been widely promoted as the future of online shopping, the study shows that functional value, trust, and psychological structure matter more than spectacle alone. 

Personalized AI and AR immersion work as a single psychological system

Much of the existing research treats AI as a background system that improves recommendations, while augmented reality is framed as a front-end feature that enhances experience. This separation no longer reflects how leading platforms operate.

The study shows that in modern e-commerce environments, AI and AR form a tightly coupled system. Personalized recommendations generated by AI directly influence how immersive AR experiences feel. At the same time, immersive AR interactions generate richer behavioral data that feed back into AI systems, improving recommendation accuracy. This bidirectional relationship creates what the authors describe as a stable psychological structure rather than a sequence of isolated effects.

From the consumer’s perspective, AI-driven personalization acts as the primary stimulus. When users receive recommendations that closely match their preferences, they experience two simultaneous psychological effects. First, they perceive the platform as more useful, because it reduces search effort and improves decision quality. Second, they become more immersed in the shopping experience, losing awareness of their physical surroundings and focusing attention on the digital environment.

Crucially, these two states are positively linked. Perceived usefulness and immersion rise together rather than operating independently. Consumers who believe the system helps them shop more effectively are more likely to engage deeply with AR features. Likewise, immersive interactions reinforce beliefs about the system’s practical value by providing sensory confirmation of product attributes and fit.

This finding reframes the role of AR in e-commerce. Immersion is not just about entertainment or visual appeal. It functions as part of a broader cognitive–experiential system that supports decision making. The study shows that the most effective platforms are those where AI-driven relevance and AR-enabled immersion are aligned rather than treated as separate innovation tracks.

Why usefulness outweighs immersion in shaping emotional response

In popular discourse, immersive experiences are often assumed to be the primary source of positive emotions in digital retail. The data tells a more nuanced story.

Both immersion and perceived usefulness generate positive emotional responses, which then increase consumers’ intention to continue using the platform. However, perceived usefulness has a significantly stronger effect than immersion. Even in AR-rich environments, emotions such as satisfaction, excitement, and confidence are driven more by cognitive evaluations of usefulness than by sensory experience alone.

This challenges the assumption that deeper immersion automatically leads to stronger emotional engagement. Instead, the study shows that emotions in integrated AI–AR shopping environments are grounded in task success. Consumers feel better when the system helps them make good decisions efficiently, not simply when the experience is visually compelling.

The implications for platform design are significant. Investment in high-fidelity AR experiences without corresponding improvements in recommendation accuracy and decision support may deliver diminishing returns. Users value immersion, but they value clarity, relevance, and efficiency more.

The study also finds that immersion does not amplify the emotional impact of personalized recommendations in real time. In other words, being deeply immersed does not make AI recommendations emotionally stronger. This suggests that current AI–AR integration emphasizes the formation of a stable psychological foundation rather than dynamic emotional amplification. Consumers gradually develop confidence and attachment to the platform through repeated successful interactions rather than momentary emotional spikes.

This structural rather than reactive integration helps explain why some AR features fail to increase long-term engagement. Without strong cognitive value, immersive experiences may feel impressive but forgettable. The research indicates that sustainable engagement depends on aligning immersive design with clear functional benefits.

Trust determines whether AI and AR actually work

Trust emerges as the critical condition that determines whether integrated technologies succeed. Rather than acting as a direct driver of behavior, trust functions as a powerful moderator. It strengthens the impact of personalized recommendations on both perceived usefulness and immersion.

When users trust a platform, they are more willing to accept AI-generated recommendations and engage deeply with AR features. Trust reduces psychological resistance and perceived risk, allowing consumers to process system outputs more openly. In contrast, low trust weakens the effectiveness of even highly personalized or immersive systems.

Importantly, the study shows that trust does not directly increase immersion or usefulness on its own. Its role is catalytic rather than causal. Trust amplifies the efficiency with which technological stimuli are converted into psychological states. This finding refines trust transfer theory by demonstrating that in mature digital platforms, trust acts as a boundary condition rather than a simple antecedent.

The research also clarifies how trust operates in integrated systems. Consumers do not separately evaluate AI and AR components. Instead, they form an overall judgment of the platform’s reliability based on repeated interactions, brand reputation, and perceived data protection. This aggregated trust then transfers to embedded technologies.

Positive experiences with one component reinforce trust in others. Accurate recommendations increase confidence in AR visualizations, while realistic AR experiences reinforce belief in AI competence. Over time, this reciprocal reinforcement stabilizes user trust and deepens engagement.

The absence of trust has the opposite effect. Even well-designed AI–AR systems struggle to influence behavior if users fear data misuse, manipulation, or bias. The study highlights that transparency and expectation management are not optional extras. They are prerequisites for making advanced technologies effective.