AI shopping recommendations were linked to impulse buying and low-need purchases

A study of 642 Vietnamese online shoppers found that trust in recommendations and emotional excitement were associated with stronger urges to buy impulsively.

Editorial illustration — not from the study.

The study examined how features of AI-mediated online shopping, including interactions with streamers and peers, perceived realness and vividness, affect shoppers’ responses. Researchers assessed whether these cues shaped trust, perceived usefulness, emotional arousal and other perceptions related to impulse buying.

Impulse buying urge was associated with perceived usefulness, trust in recommendations and arousal. The urge was then positively associated with unsustainable consumption, suggesting a behavioral pathway from AI-enabled shopping experiences to purchases that shoppers may not need. Perceived ease of use, uncertainty and perceived manipulation did not show significant effects in the analysis.

What drove impulse buying

Using data from 642 Vietnamese online shoppers and partial least squares structural equation modeling, the researchers found that impulse buying urge was significantly associated with perceived usefulness, trust in recommendations and arousal. Perceived ease of use, uncertainty and perceived manipulation were not significant predictors in the model. Impulse buying urge was positively associated with unsustainable consumption, described in the study as low-necessity purchasing.

Why the link matters

The findings identify trust in recommendations and emotional activation as important parts of the link between AI-guided shopping and unsustainable consumption. The researchers suggest that retailers and AI platform designers use more responsible recommendations, make personalization cues clearer and add opportunities for shoppers to pause and reconsider purchases.

Evidence and caveats

This was a study of 642 Vietnamese online shoppers analyzed with a statistical model of relationships among shopping features, perceptions, impulse buying and consumption. The results show associations within this sample and model; the abstract does not describe a randomized test, so they do not by themselves establish that algorithmic targeting causes unsustainable purchases. The findings may also not apply to shoppers in other countries or settings.

High-mounted cameras spot pedestrians on the device at busy intersections

Researchers introduced a 13,314-image dataset for testing AI systems that view urban intersections from above. The results show that detection can run on edge devices such as the Jetson Orin Nano, although pedestrians remain harder to detect than vehicles and performance changes with scene conditions.

4 hours ago·2 views

AI & Computing

Where smaller AI models can work alongside—and compete with—larger ones

A survey examines the roles smaller AI models can play alongside large language models. It focuses on both collaboration and competition, including how smaller systems may help researchers and businesses use computing resources more efficiently.

21 hours ago·6 views

AI & Computing

AI watermarks in Europe do not yet meet all four legal tests

The European Union’s AI Act calls for providers to mark and detect outputs from general-purpose AI models. A new analysis finds that no current watermarking approach for large language models meets all four of the law’s stated standards: reliability, interoperability, effectiveness and robustness.

2 days ago·17 views