The role of social presence in moderating the impact of artificial intelligence on perceived value in social commerce

Social commerce (s-commerce) is a combination of online shopping and social media that aims to support transactions and change the way consumers interact globally (Elshaer et al., 2024). As a digital channel, s-commerce has proven influential in disseminating information that can shape consumer attitudes and decisions (Xu et al., 2023), as it provides a space for intensive interaction between customers (Wang et al., 2021). The effectiveness of interactions in s-commerce is increasingly enhanced by the support of artificial intelligence (AI) technology, which can recommend content personally, in real time and according to user preferences. This has a positive impact on increasing product purchase conversions (Grover and Arora, 2025). An example of successful AI implementation can be seen in Amazon, which has automated various aspects of its business through chatbot services, targeted product recommendation systems and supply chain optimization (Yin and Qiu, 2021).

Generally, the use of AI in s-commerce can be viewed from two perspectives: business owners and customers. For business owners, AI is utilized in the form of chatbots, automated product description generation, data analysis, sales predictions, after-sales service and fake review detection. Meanwhile, for customers, AI provides convenience through features such as interactive chatbots, product recommendations, image recognition, personalized shopping experiences and streamlined payment processes (Febrian, 2025; Wang et al., 2023). These various AI-based systems have been widely implemented in the s-commerce ecosystem (Păuceanu et al., 2023), strengthening s-commerce's position as a key innovation in today's digital world. Therefore, to ensure long-term effectiveness and relevance, it is crucial to evaluate the performance of the AI ​​systems used regularly. This evaluation aims not only to identify weaknesses or errors in the system but also to encourage continuous improvement, adapt to dynamic user needs and improve overall service quality.

Previous research has discussed how to evaluate AI utilization from a customer experience perspective (Cheng and Wang, 2025). One approach is to use the technology acceptance model (TAM) concept (perceived usefulness and ease of use) to understand the effectiveness of AI use in online businesses (Bui et al., 2025; Wang et al., 2023). The TAM model has advantages over other technology acceptance models in the adoption of internet-based and mobile technologies because it provides a higher level of explanatory power (Ibrahim et al., 2024). TAM can increase perceived value because the perceived ease and usefulness of AI can encourage users to assess the technology as more valuable and worthwhile (Yin and Qiu, 2021). However, previous studies have overlooked the crucial role of social influence, which is a key difference between e-commerce and s-commerce. E-commerce focuses on online buying and selling transactions without involving intense social interaction, while s-commerce emphasizes the importance of social connectivity in the purchasing process (Barbosa and Santos, 2023). A lack of social presence can hinder online businesses due to the absence of human interaction and a lack of trust (Alnoor et al., 2022).

Based on this gap, this study makes a novel contribution by proposing a model that examines the moderating role of social presence in AI adoption in s-commerce. While previous research has only positioned social presence as an antecedent or direct mediator (Al-Oraini, 2025; Hong et al., 2024), its contingent role in enhancing the perceived value generated by AI has received less empirical attention, particularly in managing AI in s-commerce, which is increasingly replacing human interaction. Social presence does not necessarily explain how perceived value leads to satisfaction (Ye et al., 2026), but rather determines when and under what conditions its influence becomes stronger or weaker. Perceived value derived from functional efficiency and personalization does not automatically translate into customer satisfaction because digital interactions lack human warmth (Wang et al., 2026). Social presence acts as a condition that enhances the emotional interpretation of perceived value (Grewal et al., 2020), thereby influencing customer satisfaction. Therefore, this study examines how social presence moderates the relationship between perceived value and customer satisfaction in the context of s-commerce using AI technology. This study also provides a more comprehensive understanding of the dynamics of social interactions on s-commerce platforms and how AI can enhance customer-perceived value when supported by a strong social presence.

This research consists of several systematic stages. After identifying the research problem and conceptual framework, the research instrument was developed, and data were collected empirically. Data analysis used a tailored method and concluded with a discussion of the results.

AI is a scientific field that develops intelligent machines that mimic human cognition to learn and solve problems, and it is applied through natural language in web search, speech and images (Na et al., 2022). To sustain this technology, proper evaluation of its use is essential. The TAM is proposed as a powerful tool to explain the factors that lead users to adopt new technologies, with two key factors: perceived ease of use and perceived usefulness (Davis, 1989). Perceived ease of use is a factor that encompasses how potential users evaluate an application and refers to the ease of use of the system (Venkatesh et al., 2003). Another factor is perceived usefulness, the consumer's perception of how useful a product or service is. These two dimensions are often used to interpret perceived value (Jo, 2022). Perceived value is defined as the perceived difference between what consumers pay and what they receive during the shopping process (Zeithaml, 1988). S-commerce has successfully shifted AI's role from a functional technology to an interactive social actor (Hsu et al., 2026; Liao, 2026). Evaluation of AI systems is not solely based on their usability but also on the quality of the interactions they deliver. For example, chatbots or automated services that help interactively recommend products. The concept of anthropomorphism can also be used to examine how users tend to attribute human characteristics to AI systems (Greilich et al., 2025). For example, natural language communication, personalized responses and emotional expressions can increase consumer emotional engagement (Chaturvedi et al., 2025).

Previous research has demonstrated the importance of understanding consumer behaviour to meet their desires in online marketplaces. AI-driven behaviours can influence consumers' perceived utility and hedonic value, ultimately increasing purchase intentions (Dixit et al., 2025). However, these results require further discussion in investigating consumers' emotional responses to AI interactions. While AI has been shown to increase enjoyment and engagement during online shopping (Elmashhara et al., 2024; Febrian, 2023; Hoyer et al., 2020), highly automated interactions can actually reduce emotional intimacy and create impersonal experiences. These inconsistencies suggest that the effectiveness of AI in s-commerce depends not only on technological performance but also on users' perceptions of the social interaction during communication.

Perceived value is an essential concept in marketing that helps companies understand customer purchasing decisions and improve products or services to meet their needs (Zhang et al., 2024b). This concept can be explained through the theory of consumption value (Sheth et al., 1991), which states that consumer decisions are influenced by five value dimensions: functional value, emotional and/or hedonic value, social value, epistemic value and conditional value. However, in the context of s-commerce, the most dominant dimensions are utilitarian value and hedonic value (Lin et al., 2020). Perceived value and customer satisfaction are closely related, forming the basis for creating successful business dynamics (Yum and Kim, 2024). The basis for its formation is perceived value, which serves as a reference for customer perceptions of a product (Blut et al., 2024). When customers perceive high value, they tend to be more satisfied with their purchases (Tedja et al., 2024). Perceived value can be formed from utility and hedonic value (Yin and Qiu, 2021). The practical benefits and utility value consumers derive from a product are referred to as utility value (Zhang et al., 2024b). Customers tend to feel satisfied and inspired to make a purchase when the product displayed can meet their needs or concerns. Meanwhile, hedonic value is defined as feelings of pleasure, relaxation, excitement, curiosity, surprise and mental engagement and interest during the interactive process (Chen et al., 2025). These feelings arise from the perceived usefulness obtained through purchasing and using a brand, which evokes emotions and affection, derived from personal pleasure and satisfaction (Schnebelen and Bruhn, 2018). Increasing utilitarian and hedonic value can increase customer satisfaction (Yum and Kim, 2024). Although in some situations, utilitarian value is the primary driver of consumer satisfaction due to the immediate benefits experienced when shopping online (Liu et al., 2020), research also confirms that hedonic value plays the most important role due to its entertaining and interactive nature (Mutya and Ilankadhir, 2025). These differences in results may be due to differences in consumer experience and platform characteristics. Therefore, we hypothesize that:

Customer satisfaction is often used as a key indicator of businesses to assess the extent to which a product meets customer expectations (Nilashi et al., 2023). This evaluation is essential because it influences overall purchasing behaviour and can accurately predict future purchasing behaviour (Acar et al., 2024). User satisfaction and continued behaviour can be studied by understanding the concept of expectation confirmation theory (ECT) (Oliver, 1980). This concept compares the extent to which expectations are met before and after a purchase, thus forming satisfaction with a product or service (Yi et al., 2024). One consumer activity anticipated by marketers is repurchase intention. Repurchase intention refers to consumers' willingness to make planned purchases in the future, reflecting their belief or attitude to repurchase a particular product or service based on an evaluation of previous experiences (Febrian et al., 2025; Hussain et al., 2024). Repurchase intention can be formed if consumers feel satisfied (Nuralam et al., 2024). Therefore, marketers need to understand how to create customer satisfaction to increase repurchase intention.

The concept of social presence is used to explain user intentions in s-commerce activities and the role of social presence in the interaction process (Tao et al., 2024). The concept was introduced by (Short, 1976) to help increase trust, build relationships and encourage purchase intentions because consumers feel they are interacting with real people, not just systems. The concept explains that easy access to social presence is a crucial factor in creating effective conversations and engagement between two individuals through the support of modern technology (Kim et al., 2016). Social presence has been confirmed to stimulate positive psychological responses, such as feelings of trust, and behavioural responses, such as increased purchase intentions in both offline and online retail (Grewal et al., 2020). This can be achieved by fostering a sense of social warmth and incorporating hedonic motivation into online shopping experiences (Alnoor et al., 2022).

The emergence of human-like AI technology in s-commerce has increased consumers' sense of presence as social actors (Lee and Oh, 2021). This technology has given rise to critical discourse regarding the authenticity of the resulting social interactions (Kim and Wang, 2024; Vo et al., 2025). Although AI can mimic humans, social presence is still necessary because AI is inherently simulative and not authentically relational (Beerends and Aydin, 2025). For example, Amazon's voice assistant service features show that users are more satisfied when they can interact socially with it (Purington et al., 2017). This perception may be an essential factor preceding users' feelings of social presence with AI-powered services (Jin and Youn, 2023). Previous research has found that how users evaluate technology is determined by their perceptions of social presence (Pitardi and Marriott, 2021). Therefore, this can be used to predict user satisfaction (Hayashi et al., 2004). Although several studies have explored the underlying motivations for AI use, little is known about how various social presence factors relate to user satisfaction. This study reaffirms the critical role of social presence in users' evaluations of AI use in s-commerce. Shao and Kwon (2021) emphasize social presence as a determinant of user satisfaction. Thus, companies need to understand the role of social presence in creating more personal and meaningful interaction experiences, thereby strengthening perceived value (utility value and hedonic value) and increasing customer satisfaction.

Previous research also suggests that cultural context plays a significant role in shaping user responses to AI interactions (Lim et al., 2021). Users from individualistic cultures tend to emphasize functional efficiency over relational cues. In contrast, research in collectivistic cultures, including Southeast Asian contexts, highlights the importance of social presence and perceived social connectedness in evaluating technology (Dang and Li, 2026; Kitayama and Salvador, 2024). For example, Indonesia is characterized as a collectivist society that values ​​relational closeness (Martina et al., 2023). They value relational interaction cues and social warmth over functional performance. AI interactions that can create a stronger sense of social presence can enhance users' emotional well-being (Zhang et al., 2024a). Therefore, the role of social presence in enhancing perceived value and user satisfaction observed in this study is consistent with findings from other collectivist cultural contexts and extends the literature by providing empirical evidence from an emerging market environment.

The proposed research model, which describes the indirect influence of AI on repurchase intention and the moderating role of social presence, is illustrated in Figure 1.

The quantitative method used was to distribute questionnaires to respondents. Purposive sampling was used to target respondents who were social media users and experienced with AI features. However, this approach may limit the generalizability of the findings, as the sample may not fully represent all s-commerce users in Indonesia. Data were collected via a Google Form distributed via WhatsApp and social media. A total of 309 samples were collected with several defined demographic characteristics. Respondents' responses focused solely on questionnaire questions related to their experiences, without requesting personal data to maintain data confidentiality. Table A1 ^[1] presents the characteristics of the respondents in the survey. Based on gender, female respondents predominated at 62%. By age, young people aged 18–28 were the largest group. The dominance of young respondents can also introduce demographic bias because they are more familiar with AI technology. Based on the selected social media type, TikTok was the most popular at 40%. This figure indicates that the younger generation uses TikTok extensively.

A five-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree), was used to measure respondents' responses to the questions posed for each variable. Question items were adopted from relevant previous research. AI is conceptualized as a second-order reflective construct consisting of two dimensions: perceived usefulness and perceived ease of use, following the TAM (Shoukat et al., 2025; Venkatesh et al., 2003), social presence (Toader et al., 2020), perceived value, which has two dimensions: hedonic and utility value (Yin and Qiu, 2021), satisfaction (Sheikh et al., 2019) and repurchase intention (Shang and Bao, 2020).

This study analyzed data and tested hypotheses using a structural equation modelling (SEM) approach, conducted in two stages: testing the measurement model and the structural model. The AMOS statistical tool was used to perform these tests, including validity, reliability and hypothesis testing. SEM was chosen due to its superior statistical integrity compared to other approaches, such as multiple regression analysis (Hair et al., 2021). SEM can simultaneously test complex relationships, accommodate latent variables and consider measurement error, resulting in more accurate estimates and a model that better fits the empirical data.

Before testing the measurement model, data normality was checked to ensure that there were no missing values ​​or outliers. Data normality was checked using a z-score reference value of ±3 (Andrade, 2021). Common method bias testing was also conducted to measure errors caused by the same measurement environment and data source. To test for common method bias, this study used Harman's single-factor test to analyze all key variable items (Podsakoff et al., 2003). The amount of variation explained by the first factor was 26%, which is below the standard threshold of 40%. This indicates that there is no serious common method bias in this study, and the data can be analyzed in the next step. After the data were deemed to meet the normality requirements, the measurement model was then tested using AMOS. Confirmatory factor analysis was used to test and ensure the suitability of the measurement model through validity and reliability. Several methods can be used to test validity and reliability. Convergent validity is seen from the loading factor value of all question items >0.5 and the average variance extracted > 0.5 (Hair et al., 2020) shown in Table 1. Thus, all constructs have the recommended validity value. Furthermore, reliability is seen from the composite reliability (CR) >0.7 as recommended (Nunnally and Bernstein, 1994; Sarstedt et al., 2022). The CR value of all constructs is above 0.7. All these results confirm the validity and reliability criteria according to the required values, so that it can be continued to the next stage, namely, testing the hypothesis or structural model. Goodness of fit testing on the measurement model is also seen in chi-square (χ²/df) = 2.410, goodness of fit index (GFI) = 0.838, incremental fit index (IFI) = 0.924, normed fit index (NFI) = 0.877, Tucker–Lewis index (TLI) = 0.912 and root mean square error of approximation (RMSEA) = 0.068. Overall, the measurement model demonstrated an acceptable to marginally acceptable fit according to commonly recommended SEM fit criteria.

The path diagram for the research model explains the unstandardized path coefficients and decisions for each hypothesis. Table A2 ^[1] presents both standardized path coefficients and significance values obtained from the SEM analysis. Hypothesis 1 confirmed that AI has a significant positive effect on utility value (β = 0.915, p = <0.001), indicating that users who perceive AI features as useful and easy to use tend to perceive significantly greater utilitarian benefits in s-commerce. Hypothesis 2 also showed that AI has a significant positive effect on hedonic value (β = 0.916, p = <0.001), indicating that AI-driven features significantly enhance users' enjoyment and emotional shopping experience. Hypothesis 3 confirmed that utility value has a significant positive effect on satisfaction (β = 0.308, p = <0.001). This finding suggests that practical benefits such as efficiency and convenience contribute to customer satisfaction. Hypothesis 4 indicated that hedonic value has a slightly stronger, moderate effect on satisfaction (β = 0.388, p < 0.001), implying that emotional and enjoyable shopping experiences play a more dominant role in shaping satisfaction than purely functional benefits. In addition, satisfaction has a significant positive effect on repurchase intention (β = 0.970, p = <0.001). Thus, Hypothesis 5 is confirmed. These results indicate that satisfied users are highly likely to continue purchasing through the same s-commerce platform. The structural model demonstrated an acceptable level of fit based on the overall GFI indices through chi-square (χ²/df) = 2.923, GFI = 0.841, IFI = 0.919, NFI = 0.881, TLI = 0.905 and RMSEA = 0.079.

Moderation analysis was conducted to examine whether social presence moderated the relationships between utility value and satisfaction, as well as between hedonic value and satisfaction. The moderation effect was tested using the product indicator approach in SEM. The indicators of utility value, hedonic value and social presence were mean-centred to reduce potential multicollinearity. Two interaction constructs were created by multiplying the centred indicators of utility value and hedonic value by social presence. These interaction constructs were then incorporated into the SEM model to assess the moderating role of social presence. The results showed that the interaction effects of social presence were positive and statistically significant for the utility value–satisfaction relationship (β = 0.366, p = 0.002) and the hedonic value–satisfaction relationship (β = 0.277, p = 0.004). The positive interaction coefficients indicate that higher levels of social presence strengthen the positive effects of utility value and hedonic value on customer satisfaction. The empirical results indicate that Hypothesis 6 and Hypothesis 7 are supported. To further interpret the moderating effects, this study employed Jeremy Dawson's simple slope analysis (Dawson, 2023), comparing the relationships at low (−1 standard deviation (SD)) and high (+1 SD) levels of social presence. Figures A1 ^[1] and A2 ^[1] show that the relationships between perceived value dimensions and satisfaction are stronger at higher levels of social presence. The steeper slopes under high social presence indicate an amplifying moderating effect.

This study investigates the moderating parameters of social presence, including a performance evaluation of AI utilization to increase repurchase intention in s-commerce. Our study found that AI utilization, shaped by perceived ease of use and usefulness, can increase perceived value in the form of hedonic and utilitarian values. These results support previous research explaining that high usefulness and ease of use of AI can increase hedonic and utility value (Noerman et al., 2025; Xie et al., 2024). These results confirm that the adoption of AI-based technology is not only viewed as a tool to increase efficiency but also as a means of providing enjoyable experiences for consumers through more personalized and relevant interactions. When consumers perceive AI as easy to use and useful, they not only experience functional value (utility) such as time savings and increased productivity but also derive hedonic value in the form of satisfaction, pleasure and a more engaging experience. This is consistent with the concept of the TAM and the theory of consumption value perspective, which emphasizes that perceived usefulness and ease of use will drive the formation of more comprehensive consumer value, encompassing both rational and emotional aspects.

The results of this study also confirm that social presence can strengthen the influence of perceived value, both hedonic and utilitarian, on increasing customer satisfaction. These results provide new evidence that social presence plays a crucial role as a catalyst in strengthening the influence of perceived value, both hedonic and utilitarian, on satisfaction. Social presence provides a sense of social warmth and fuels hedonic motivation for online shopping (Alnoor et al., 2022). A strong social presence makes consumers feel closer, more connected and more valued in online interactions, making the perceived value of a product or service more easily translated into satisfaction. Without social presence, perceived value may be understood solely as functional benefits, but with real social interaction, this value expands into the emotional and experiential realms, thus having a greater impact on satisfaction. Therefore, the theory of consumption value is validated and confirmed in this study. Furthermore, this study creates an empirically validated model that contributes to the body of knowledge in s-commerce.

This study also provides insight into why social presence strengthens the effect of perceived value on customer satisfaction. This sense of social presence makes users feel cared for and understood, making the interaction experience more meaningful (Hamilton et al., 2021). Furthermore, social presence also increases trust because users perceive autonomy or control in the interaction process, thus strengthening their confidence in the system they are interacting with (Pavone and Desveaud, 2025). When AI interactions provide human-like responsiveness, empathy and contextual understanding, consumers are more likely to perceive them as authentic rather than merely algorithmic. This perceived authenticity reduces psychological distance and increases consumer trust in the platform. Thus, increased social presence fosters emotional engagement by making consumers feel socially validated and personally involved during the shopping process. These findings extend social presence theory, which posits that AI-based interactions can increase satisfaction by fostering perceived authenticity. Perceived value is strengthened when AI interactions are perceived as socially authentic and trustworthy rather than simply impersonal technological interfaces.

This study contributes theoretically by demonstrating that the effectiveness of AI in s-commerce is shaped not only by technological perceptions such as usefulness and ease of use but also by socially embedded interaction experiences, as reflected in social presence. While previous studies mainly examined AI from functional and technological perspectives, this study shows that social presence strengthens the transformation of utilitarian and hedonic value into customer satisfaction. These findings provide a more integrated understanding of how technological and social factors jointly influence repurchase intention in an AI-based s-commerce environment. Ultimately, this research confirms that increasing customer satisfaction through the use of AI in s-commerce can increase repurchase intention. These results support previous research (Dhaigude and Mohan, 2023; Hui et al., 2025), which asserted that customer satisfaction plays a key role in building repurchase intentions, mainly when supported by interactive experiences utilizing digital technology (Loke et al., 2025). This research confirms the concept of ECT, which emphasizes satisfaction as a key variable that bridges the match between initial expectations and actual experiences. In the context of AI-based s-commerce, customers not only assess the suitability of product or service functions but also associate the digital interaction experience with their level of satisfaction. When satisfaction is achieved, expectation confirmation becomes stronger and encourages consumers to make repeat purchases. Thus, this research extends the application of ECT to the modern digital context by demonstrating that the use of AI can strengthen the relationship between customer satisfaction and repurchase intentions.

The results of this study are helpful for practitioners in evaluating the effectiveness of AI in shaping positive customer perceptions, which can serve as a basis for future marketing strategies. The use of AI in s-commerce aims not only to increase efficiency but also to create enjoyable and relevant shopping experiences for consumers. Businesses can integrate easy-to-use, useful AI technology to strengthen personal interactions, thereby enhancing perceived value from both utilitarian and hedonic perspectives. Thus, AI not only provides functional value in the form of time savings, efficiency and increased productivity but also provides satisfaction, enjoyment and an engaging shopping experience for consumers. The implementation of easy-to-use and useful AI features may help enhance customer perceptions of value and satisfaction during s-commerce interactions.

Furthermore, this study's findings emphasize the importance of establishing and strengthening a social presence in online interactions when using AI. A tangible social presence, such as live chat, interactive reviews, real-time customer service or virtual communities (Grover et al., 2026), can strengthen consumers' perceptions of both the utilitarian and hedonic value of a product or service. This is important because consumers assess not only functional benefits such as efficiency and usability but also seek emotional experiences such as a sense of appreciation, closeness and connection with the seller and other user communities. In practice, companies can implement personalized AI chatbots to greet consumers by name or provide recommendations based on shopping history, introduce live shopping features that enable real-time interaction between sellers and consumers and build virtual communities that serve as a platform for consumers to share experiences. Furthermore, companies can respond to reviews interactively with personalized replies and use AI-based sentiment analysis to gain a deeper understanding of consumer perceptions. With these steps, social presence will be strengthened, allowing consumers to experience not only functional benefits but also emotional experiences.

Since AI significantly influences utilitarian and hedonic value, platforms should focus on AI features that improve usability, responsiveness and recommendation relevance during shopping interactions. Companies can use AI interaction data to better understand customer preferences and improve shopping experiences. With this technology, companies can better understand consumer behaviour patterns, anticipate obstacles that could reduce satisfaction and design proactive strategies to maintain a consistent customer experience at every stage of the shopping journey. Furthermore, implementing AI-based co-creation platforms is crucial because they allow consumers to participate in product development, campaign design or service experiences through feedback intelligently analyzed by AI. This may help increase customer satisfaction and encourage repeat purchase behaviour.

By achieving satisfaction, consumers will feel their expectations are confirmed, ultimately strengthening loyalty and the likelihood of repeat purchases. This demonstrates that implementing AI not only improves operational efficiency but also serves as a strategic tool for building long-term customer relationships. Companies can leverage AI to develop personalized product recommendation systems tailored to consumers' preferences and shopping history. In this way, AI can be used to provide product recommendations based on consumers' shopping preferences and previous interactions. Implementing activities such as integrating innovative notification features to remind consumers of promotions or products that align with their interests, and using AI-based data analysis to identify the most frequent points of satisfaction and complaints, can be beneficial. These steps can continuously improve customer satisfaction, ultimately strengthening loyalty and encouraging repeat purchases.

This study concludes that the use of AI in s-commerce, driven by perceived ease of use and perceived usefulness, increases perceived value across utilitarian and hedonic dimensions, thereby boosting customer satisfaction and ultimately increasing repurchase intention. Social presence has been shown to moderate this relationship by strengthening the emotional value and experience of interactions, so that satisfaction is formed not only from functional benefits but also from a sense of closeness, appreciation and connectedness among consumers, sellers and online communities. These results confirm the relevance of the TAM, the theory of consumption value and ECT in the modern digital context, demonstrating that satisfaction plays a key role in linking technological benefits to customer loyalty. Practically, this study emphasizes the need for companies to integrate AI not only for efficiency but also to create personalized, interactive and relevant shopping experiences. Companies are advised to build AI-driven customer journey mapping that can predict consumer behaviour end-to-end and integrate AI with human touch through hybrid services, so that digital interactions remain warm and emotionally valuable. Additionally, companies can develop AI-based co-creation platforms that involve consumers in providing input on product or campaign designs, so consumers feel part of the business process. This approach not only increases satisfaction and loyalty but also creates competitive differentiation that is more difficult for competitors to replicate.

While this study makes several contributions to existing research, it has several limitations. First, the study employed a cross-sectional design, which limits the ability to infer causal relationships among AI utilization, perceived value, satisfaction and repurchase intention. Longitudinal studies are therefore recommended to examine how consumer perceptions and behaviours toward AI in s-commerce evolve. Second, the study relied on self-reported quantitative data collected from users across multiple s-commerce platforms. Although this approach provides broader generalizability, it may obscure platform-specific characteristics and interaction patterns that could influence consumer responses toward AI features. Future research could therefore focus on a single platform or compare platforms more effectively better to understand contextual differences in AI-driven s-commerce experiences. Third, the exclusive use of survey-based measures may increase the risk of common method bias and fail to capture consumers' deeper emotional and behavioural experiences fully. Future studies could integrate qualitative or experimental approaches to explore in greater depth how AI-generated interactions shape consumers' emotional and behavioural experiences. From a theoretical perspective, future research could further extend the TAM, theory of consumption value and ECT by incorporating additional psychological and relational constructs, such as trust in AI, perceived transparency, consumer engagement or anthropomorphism. Future studies may also examine whether the effectiveness of AI differs across demographic groups, cultural contexts or product categories.

The author would like to express his sincere gratitude to the reviewers and participants of the 7th Asian Conference on Business and Economic Studies (ACBES 2025) for their valuable input and constructive suggestions, which have significantly improved the quality of this manuscript, and to the Universitas Lampung for supporting the completion of this research.

The supplementary material for this article can be found online.