✨
International Knowledge Platform
Serving Researchers Since 2012

The Impact of AI-Powered Personalization on Consumer Purchase Intention

DOI : 10.17577/IJERTCONV14IS110119
Download Full-Text PDF Cite this Publication

Text Only Version

The Impact of AI-Powered Personalization on Consumer Purchase Intention

Ms. Mabel Mathias1

Master of Business Administration, Administrative Management College, Bannerghatta Road, Bengaluru

Email id: mabelmathias8971101916@gmail.com Mob. No: 8971101916

Mr. Nithin R

Master of Business Administration, Administrative Management College, Bannerghatta Road, Bengaluru

Email id: nithin4481@gmail.com Mobile no: 8861676247

Guided by Mrs. S. Gandhimathi, Assistant Professor,

Department of Management Studies(MBA)

Administrative Management College, Bengaluru

Email id: sgandhimathimcs@gmail.com Mob No: 7291857123

Abstract – Artificial intelligence (AI) has become an important driver of digital transformation, changing how businesses understand consumers, design marketing communication, and support online purchase decisions. AI-powered personalization uses consumer data, machine-learning techniques, recommendation systems, predictive analytics, and automated content tools to provide products, offers, and messages that are more relevant to individual needs. This paper examines how AI-powered personalization may influence consumer purchase intention using exclusively secondary data. A descriptive and analytical research design is adopted, drawing on peer-reviewed literature, established theoretical models, industry research, and responsible-AI guidance. Four objectives guide the study: to identify major AI personalization practices; to examine the mechanisms linking personalization with purchase intention; to analyse the roles of trust, privacy, transparency, and perceived intrusiveness; and to develop practical recommendations for responsible implementation. The literature indicates that relevant personalization can reduce information-search effort, improve perceived usefulness, increase engagement, and make purchase decisions easier. However, personalization does not automatically produce positive outcomes. Consumers will also have a negative response to situations where information gathering seems surreptitious, recommendation seems manipulative, or decisions are hard to comprehend. The relationship is therefore conditional: relevance and convenience can encourage purchase intention, while privacy concerns and low trust can weaken or reverse the effect. The paper proposes an integrated conceptual framework combining the Technology Acceptance Model, the Theory of Planned Behaviour, and the personalizationprivacy paradox. It concludes that organizations should evaluate personalization not only by clicks and conversions but also by consent quality, consumer control, fairness, transparency, and long-term trust. Because the study relies on secondary evidence, its findings are interpretive and do not represent new survey-based causal estimates.

Keywords: Artificial intelligence; AI-powered personalization; consumer purchase intention; digital transformation; consumer trust; privacy; recommendation systems.

  1. INTRODUCTION

    Digital transformation refers to the integration of digital technologies into business processes, customer relationships, and business models. Artificial intelligence has enhanced the ability of organizations in the marketing field to analyze volumes of behavioral data and tailor digital experiences at scale. E-commerce websites, streaming companies, travel agents, financial institutions, and retail stores can leverage AI to recommend products, order search results, provide offers, anticipate customer needs, power conversation

    bots, and personalize messages targeting specific categories of consumers. In this way, businesses could make the customer experience more timely and relevant and coordinate communication through websites, mobile phones, emails, and other channels.

    The economic explanation behind these marketing techniques is that consumers are overloaded with both product offers and messages. By offering relevant recommendations, companies can save consumers' time searching for products and introduce them to the appropriate alternatives. According to research conducted by the McKinsey & Company (2021), 71% of consumers surveyed expected personalized communication from the company and 76% stated that they felt frustrated when it was missing. This shows that personalization is becoming one of the key customer-experience expectations, although results of industry surveys cannot be considered universal estimates. The advantages of AI personalization also need to be taken into account along with the risks. The fact that personalization relies on data and inferences which the consumers may not anticipate is also one of the aspects that needs to be addressed. According to Aguirre et al. (2015), this is referred to as the personalization paradox: although personalization enhances relevancy, it might simultaneously lead to increased vulnerability and diminished effectiveness of advertisements because the consumers do not perceive that their data was gathered with enough consent. Trust, privacy, control and transparency become part of the mechanism of influence on the consumers' reactions.

  2. FORMULATION OF THE PROBLEM

    Organizations gain access to AI mechanisms that allow them to discern consumer preferences and adapt digital communication to suit those preferences. However, technological skills do not automatically lead to the approval of the consumer. Institutions of any kind have to decide what information is utilized for forecasting, what degree of personalization should be addressed, when to state that the AI had an impact, and by what measure the results of their actions should be evaluated. The dilemma at hand thus involves the trade-off between the benefits of the relevance and convenience on one side and risks of privacy concerns, manipulation, bias and lack of trust on the other. Studies have focused on the various aspects of the relationship between them.. These strands need to be considered together to understand why personalization encourages purchase intention in some circumstances but has a weak or negative effect in others.

  3. STATEMENT OF THE PROBLEM

    The central research question is: What is the impact of AI-driven personalization on consumer behavior, and under what circumstances can personalization affect consumer purchase intentions positively or negatively? Specifically, the company needs to be able to answer such questions as: Will personalization lead to increased perceived relevance and ease of decision-making? What is the role of trust and privacy in determining consumer response to personalization? How can ethical considerations help to maintain consumer relationships in the age of digital disruption? The matter of this topic is of great importance because the company can invest significantly into the short-term gains of utilizing personalized AI while undermining the potential to build lasting trust and consumer engagement.

  4. RESEARCH GAP

    The authors thought that the prior research is important, but it is fragmented, addressing different aspects of personalization, online advertising, technology acceptance, and the privacy issue. First, some articles address personalization as a marketing tool, whereas others address consumer attitudes toward AI or its acceptance. This division creates a gap in understanding the entire chain from personalization algorithms to their evaluation by consumers and the impact on buying intentions.

    Second, personalization is often framed as a positive phenomenon, whereas the compensation between privacy and personalization is addressed separately. Aguirre et al. (2015) demonstrate how practices of data collectin impact the effectiveness of targeted

    advertising; therefore, one cannot address this issue without considering the interplay of relevance and value against consumer control, transparency, trust, and intrusion.

    Third, industry reports are informative but should not be used as peer-reviewed primary sources. One should understand that findings apply only to the specific audience, period, or product. One cannot generalize the observed patterns across all consumers due to differences in sensitization, literacy, and attitudinal components.

    Fourth, the topic of AI-personalization is emerging and dynamic, addressing new trends and products. The focus on the prediction paradigm and the opportunities offered by generative AI to communicate one-to-many with personalized messages should prompt organizations to find ways to govern personalization processes to respect privacy and data protection. Therefore, the paper attempts to address this gap by performing a secondary-data analysis to synthesize the prior research. The study integrates the technology acceptance model, purchase-intention framework, and personalization-paradox analysis to develop a contextual factor framework or managerial model to navigate the boundaries of relevance to consumer and the rights to privacy and trust. The statement that nobody said that personalization research has not been undertaken before is correct, but there is a need to conceptualize these findings to synthesize the body of knowledge to respond to the research question.

  5. Objectives of the Study

    The study has four objectives:

    To identify the major forms of AI-powered personalization used in digital marketing and customer-facing platforms.

    To examine how perceived relevance, usefulness, convenience, and engagement may connect AI-powered personalization with consumer purchase intention.

    To analyse how trust, privacy concern, transparency, consumer control, and perceived intrusiveness may strengthen or weaken this relationship.

    To develop practical recommendations for implementing AI-powered personalization responsibly within digital transformation strategies.

  6. Theoretical Framework

    The paper draws on three complementary perspectives: the Technology Acceptance Model, the Theory of Planned Behavior, and the personalizationprivacy paradox. Together, these perspectives explain perceived utility, intention formation, and the conditions under which personalization can become counterproductive.

    1. Technology Acceptance Model (TAM)

      Davis (1989) suggested that two factors, namely perceived usefulness and perceived ease of use, are key predictors of information technology acceptance. When applying to the acceptance of AI personalization, perceived usefulness relates to how useful the recommendations or tailored services are for consumers to find products that are likely to be suitable, compare product options and enjoy the time saving, effort reduction and better decision-making. Perceived ease of use is related to the whether or not the AI recommender is less difficult to use and does not require much effort. If the AI recommender can improve the usefulness of the product search experience while not introducing too much friction in product search, purchase intention can be built up.

    2. Theory of Planned Behavior (TPB)

      Ajzen's (1991) Theory of Planned Behavior provides an understanding of the formation of intentions through three variables, namely attitudes, subjective norms, and perceived behavioural control. With that said, personalization could influence the attitude formation by making the intended product more relevant or useful. At the same time, social approval plays a role in the subjective norms surrounding the acceptance of AI-driven practices, such as recommendation systems, ratings, and social media cues. Finally, the perception of control is vital to purchase intentions, as it shows the consumer's belief in their ability to purchase based on their income, availability, and overall convenience level. By those means, one should note that personalisation is not the sole factor in purchase intentions but rather a key part of the established system.

    3. Personalizationprivacy paradox

      The personalization-privacy paradox refers to the dilemma where consumers see advantages of relevancy but at the same time have worries about the practices of data collection. For example, consumers enjoy relevance in recommendations but object to unknown tracking, ambiguous consent or the use of sensitive information. Aguirre et al. (2015) also argue that there is an influence of data collection and trust cues on consumers' reactions to personalized online advertising, implying that relevancy and privacy cannot be assessed separately; The perceived consumer control and consumer trust may serve as moderators.

    4. Integrated conceptual framework

      The framework suggests that personalization features (such as personalized recommendation quality, personalized offers, relevant content, and predictive help) affect perceived relevance, perceived usefulness, convenience, and engagement, which in turn drive attitude and purchase intentions. Moderators such as trust, transparency, control, privacy concern, perceived intrusiveness, and perceived fairness, are likely to moderate this relationship. Contextual factors, such as product risk, consumer experience, category involvement, and the sensitivity of the data, can also influence these relationships.

      The framework is exploratory and therefore leaves room for further investigation. It is proposed as a way to organize the extant literature on consumer privacy concerns and decision-making in online settings. The framework was not statistically tested in this study.

  7. Literature Review

    1. Personalization and customer relevance

      The concept of personalization is the process through which information is adapted to consumers different needs. Tam and Ho (2006) conducted research into web personalization and how it has effects on information processing, with their findings showing that personalization affects the way users process online content. Personalized content is better aligned with a users expectations and assists in making better decisions. Also in e-commerce, personalized ranking and recommendations help to avoid unnecessary engagement in searching through the huge variety of products. Quality of personalization is dependent on the quality and quantity of information, relevance of the recommendation, and the goals of the consumer. As stated in the report by McKinsey & Company, personalized communication of the companies has been more in demand by consumers, indicating that personalization is becoming an increasingly important part of the perceived quality.. However, the report is based on the observation of managers, and does not give evidence that AI personalization results in improved performance.

    2. AI in advertising and marketing

      AI has evolved personalization from rule-based segmentation to data-driven prediction and automated content delivery. Ford et al. (2023) provide an overview and guidelines for AI advertising, a set of applications which span the advertising process rather than a singular tool. Gao et al. (2023) review developments, challenges, and ethical considerations across four areas of AI advertising: targeting, personalization, content creation, and ad optimization. The four areas closely align with the personalization practices summarized in Table 1 of this paper, which demonstrates that AI personalization is embedded across multiple stages of the customer journey, from defining potential audiences to aapting messages and measuring response.

      Content creation is the fastest-changing of these areas, with generative AI enabling the production of individualized text and images at scale, which may enable increased efficiency and better alignment with consumer interests. Gao et al. (2023) also highlight ethical considerations which come alongside these capabilities, including data use, fairness, and the transparency of automated decisions. For this study, the implication is that the same tools which increase relevance may also raise privacy and trust concerns, which further underlines the necessity to examine personalization and consumer protection together.

      Vaid et al. (2023) applied topic modelling to empirical consumer research on AI, identifying the themes through which consumer scholars have examined the technology. Their work suggests that research on AI in consumer contexts is distributed across multiple, disparate themes. This supports the integrative approach of the present paper, which binds together personalization, technology acceptance, and privacy in a singular framework. It must be noted that these three sources are largely overviews and reviews. They describe the scope, opportunities, and risks of AI in advertising, but do not independently establish a causal effect of AI personalization on purchase intention.

    3. Personalization, privacy, and trust

      Aguirre et al. (2015), as they find that, for personalized ads, there was no positive, or even negative effect, when it was discovered that the information was acquired without users being informed about it. As a result, the strategies to establish trust and the legitimacy of the collection matter.

      Some data types are more privacy sensitive (even if the consumer may not know the data collection) and consumers may have different reactions depending on how long data is stored, and shared or with third parties. Consumers may put up with data use when it's familiar and valuable (or at least reasonable), but reject it when consumers experience inappropriately familiar algorithms..

      Trust can decrease the perception of uncertainty in interactions. If consumers are confident a firm is capable, honest and with good intentions towards them, and that their personal information is in safe hands, they may be less wary of tailor-made suggestions. Trust must not be seen as a way of bypassing real protection.

    4. AI and consumer behaviour

      In research conducted by Wadhwani and Eastman (2024), the hybrid review of AI technology affects consumer behavior lays down various forms through which consumers respond negatively and positively toward AI encounters. The authors argue that the reactions of consumers toward AI are multi-layered. Consumers do not only analyze the output; they also react toward the effectiveness of the machine against judgment criteria such as whether it is comprehensible, dependable, anthropocentric, and authoritative

    5. Responsible AI and transparency

      The National Institute of Standards and Technology (NIST, 2023) identifies trustworthiness characteristics for AI systems, including validity and reliability, safety, security and resilience, accountability and transparency, explainability, privacy enhancement, and

      fairness. These principles are relevant to personalization because consumers are affected by the decisions and inferences generated by AI systems even when the underlying models are not visible to them. Governance should include continuous assessment of risks throughout the system lifecycle.

    6. Synthesis of the literature

      Through the analysis of various references, three patterns can be observed. First, personalization appears to increase the perceived relevance of offers and make the process of searching for products easier. Second, the effect of personalization on purchasing intention depends on more than just the accuracy of algorithms; various factors should also be considered, including trust, privacy, transparency, and the degree of perceived intrusion. Third, the strength of the relationship between the variables may vary in different contexts. Therefore, the review of literature allows for drawing a conditional conclusion that AI and personalization will facilitate purchase intention when helpful, accurate, reasonable, expected, and follow proper data practices.

  8. RESEARCH METHODOLOGY

    1. Research design

      The study adopts a descriptive and analytical research design based exclusively on secondary data. It describes established applications and concepts related to AI-powered personalization and analytically compares findings and arguments from prior literature. No primary data were collected, and no survey, interview, experiment, or new consumer-level dataset was used.

    2. Sources of data

      The paper draws on peer-reviewed journal articles, conceptual and review papers, an industry report, and official AI risk- management guidance. Sources were selected for their relevance to personalization, consumer behaviour, purchase intention, AI advertising, privacy, trust, and technology acceptance. Core theoretical sources were included to explain the relationship between perceived usefulness, attitudes, behavioural intention, and technology use. Industry evidence was used as context and not treated as equivalent to peer-reviewed causal findings.

    3. Method of analysis

      A narrative thematic analysis was applied to the selected sources. The material was organized around five themes: (1) AI personalization applications; (2) perceived relevance, usefulness, and convenience; (3) trust and purchase intention; (4) privacy, transparency, and intrusiveness; and (5) responsible implementation. The themes were compared to identify recurring insights, areas of tension, and implications for managers. The findings are qualitative synthesis rather than statistical meta-analysis.

    4. Scope of the study

      The study focuses on consumer-facing digital environments, especially online retail and digital advertising, while drawing on broader consumer-AI literature where relevant. It addresses purchase intention rather than actual sales alone. It does not focus on the technical construction of machine-learning models or on a single company.

    5. Data limitations

      The assessment is based on how accessible the information is, how thorough or limited it is, and how good the published sources are. The literature on the given topic examines different geographical areas, the industries analyzed, methods used in the research, and definitions of personalization, and that does not allow comparisons. Industry reports may have their own samples and measures, and as a result, we steered clear of summing the percentages from the publications considered or of finding a single value of effect. The findings should be seen as evidence-based synthesis rather than proof of causation in every context.

  9. RESULTS AND FINDINGS

    1. Forms of AI-powered personalization (Objective 1)

      The secondary sources point to six main forms of AI-powered personalization in consumer-facing marketing. Table 1 summarizes them, together with the data they typically rely on, the benefit they offer consumers, and the main risk they carry. This summary draws on the overviews of AI advertising by Ford et al. (2023) and Gao et al. (2023), and on the personalization literature discussed in Section 7.

      Table 1: Major forms of AI-powered personalization

      Form of personalization

      Typical data used

      Main consumer benefit

      Key risk

      Recommendation engines

      Browsing, purchase history, ratings, similar-user behaviour

      Easier product discovery; lower search effort

      Repetitive or inaccurate suggestions; narrowed choice

      Personalized search ranking

      Query history, location, past clicks

      Faster access to relevant products

      Opaque ranking; limited exposure to alternatives

      Customized offers and pricing

      Purchase history, loyalty data, predicted value

      More relevant promotions and discounts

      Perceived unfairness or manipulation

      Targeted advertising and email

      Behavioural, demographic, and inferred interest data

      Messages matched to current needs

      Covert tracking; sensitive inferences; reactance

      Chatbots and

      conversational assistants

      Conversation content, account and order data

      Immediate, tailored help during evaluation

      Errors; unclear disclosure that AI is in use

      Generative AI content

      Profile and context data, prompts

      Individualized text and images at scale

      Inaccurate claims; persuasive pressure; low transparency

      The secondary analysis then produced five main findings.

    2. Main findings

      Finding 1: Relevance and convenience are the primary pathways through which personalization may encourage purchase intention. AI-enabled recommendations can help consumers identify products aligned with their preferences and reduce the time required to

      The length of the impact of new technologies is longest when they output usable information. Poor recommendations may result in extra cognitive load and create a perception of incompetence.

      Finding 2: Perceived usefulness counts more than just presence of AI. Consumers are not likely to strengthen their intention to buy anything merely due to the announcement from a company that it is using artificial intelligence. The system must improve the task of shopping, i.e., find the right options, clarify the differences, and deliver necessary messages. This finding complies with the Technology Acceptance Model (Davis, 1989), which puts emphasis on perceived usefulness and ease of use and is to be assessed from the consumers angle rather than by technical sophistication.

      Finding 3: Trust and privacy are the must conditions. Personalization needs information, but customers may ask how that information has been collected. The literature related to personalization paradox states unexpected or non-consensual data collection can lead to negative effects regarding the response to tailored advertising (Aguirre et al. 2015). A relevant offer may not perform well if it results in the feeling of being under surveillance and control.

      Finding 4: Excessive or inaccurate personalization can produce negative outcomes. Repeatedly offering the same merchandise, making inaccurate conclusions on preferences, or utilizing personal data in communications can make people uncomfortable. Personalization may reduce exposure to options, sustain wrong beliefs, or make consumers feel that they are forced to buy products. These dangers imply that a system should allow users to change their preferences and supply ways for ordinary non-personalized discovery.

      Finding 5: Responsible implementation is an enabler of effective personalization, not only a compliance requirement. The ethical concerns mentioned in the AI marketing literature (Gao et al. 2023) and the features of reliability described in the NIST AI Risk Management Framework (NIST 2023) lead to the same conclusion. The concepts of clarity, explainability, and privacy protection contribute to the fact that personalization can be terms legitimate by consumers. Solutions like transparency and the guarantee of consumers giving informed yet meaningful consent, possessing the possibility of refusing to provide their data, and letting people monitor critical decisions taken by AI increase the notion of trust, which is a determining factor in purchase intention. Thus, effective governance should be integrated into personalization processes right from the beginning rather than added when the issue arises.

  10. Discussion of Findings

    The findings indicate that AI-powered personalization supports purchase intention mainly through relevance, usefulness, and convenience, while trust, privacy, and fairness determine whether these benefits are actually realized. This pattern is consistent with the integrated framework proposed in Section 6.4.

    The relationship is nevertheless not linear. Increasing the amount of data or the intensity of personalization does not necessarily improve consumer response. A recommendation that feels accurate and timely may communicate that the business understands the consumer. The same recommendation may create unease if it reveals unexpected tracking or relies on data the consumer did not knowingly provide. This finding provides support for the personalizationprivacy paradox and suggests the inclusion of trust and legitimacy in any set of determinants of purchase intentions. One implication of this study is that purchase intention is an intervening variable. A consumer may indicate purchase intention after receiving a recommendation but still decide not to buy the item due to various factors such as price, delivery terms, risk, and availability. In a similar manner, an increase in click-through rates can be an indication of interest, but it cannot be used as a reliable predictor of purchase intention. The use of behavioural indicators should be complemented by asking for opinions and measuring long-term effects.

  11. MANAGERIAL AND PRACTICAL IMPLICATIONS

    In order to operationalize personalization, one has to identify a specific customer need. As such, the company should invest in personalization powered by artificial intelligence to make the search easier, compare different options available to the consumer, and recommend products based on the consumers preferences. On the other hand, firms should avoid complicating the search process by implementing unnecessary personalization features that do not serve any purpose. Next on the agenda is maintaining high quality and relevance of data.

    Old purchase history or erroneous inference is likely to yield wrong recommendations. Companies need to ensure that customers have the opportunity to change their preferences and remove old signals whenever needed. Make the data usage clear to customers. Simplistically explain the kind of data collected, reasons for its collection, and customers ability to change their choices. Consent has to be meaningful and free of complex language.

    .

    Give consumers control over their data. You should give people the power to control how they are being personally offered; allow them the option to opt-out, reset recommendation history, and surf non-personalized content where possible. This can help combat the impression of being 'trapped'. Personalisation vs. exploration. Personalised recommendations aren't meant to be a maze of the same product, so provide alternatives, give explanations how these are generated and try to keep recommendations broad.

    Never use personal information in a manner that could psychologically or emotionally destroy the customer. Be more cautious with sensitive categories and their context before implementing machine learning. For sensitive categories, there should be stricter rules of monitoring by human. Study the impact of personalized recommendations on customer trust. It should look at purchase intent, purchase, engagement with customer, complaints relative to the number of recipients who opt ou.

  12. LIMITATIONS OF THE STUDY

    The research possesses a number of weaknesses, Firstly, the research is descriptive in that it solely utilizes secondary data and does not acquire primary input from the consumers or the businesses. Therefore, it does not provide enough evidence either of causal relationships or of the scope of the established link between personalization and the intention to purchase. Secondly, different sources adapt varied definitions, samples, sectors, and research methods that complicate the process of making a facile comparison. Thirdly, the results of the industry are connected with certain markets and time periods of surveys, meaning they should be generalized cautiously. Fourthly, AI personalization evolves at a fast pace, especially in terms of generative AI whereby practices and expectations of consumers might undergo changes after the publication. Fifthly, the narrative method might fail to include all vital studies and it does not offer systematic review or meta-analysis in terms of formal reproducibility. Finally, the intention that was not put into practice is not always considered as the actual buying activity.

  13. FUTURE DIRECTIONS

    The future work concerning the topic should guarantee the gathering of behavior data and conducting comparative studies. Such studies should take into consideration the discrepancies between intention to buy, actual purchases, and repeated purchases. It is possible to conduct experiments to check the influence of certain factors of modern recommendation solutions on such factors as perceived relevance of the recommendations, presence of AI explanations of recommended goods, and degree of users control over the process of decision making. Cross-cultural studies that will include drivers from India will find out whether digital literacy, age, language, privacy, etc., influence purchasing decisions.

  14. CONCLUSION

AI in digital transformation to have a role in more approaches to utilizing products can be encountered according to the second literature to our review, can meet the application may beused to make the 'product offer', promotion and engagement by the digital branded entities customised based on their needs. Whether the invention can be helping towards the consumer purchase intention has been considered in the paper above, should be established if it increases relevance, utility and relevant judging time and confidence in judging products.

The information also shows that there is a paradox about personalization. The data needed for an effective personalized experience could threaten privacy, if the collection or application is surprising, ambiguous, invasive or perceived to be exploitative. Trust, transparency, consumer agency and fairness are therefore drivers of why this type of personalization can be embraced or repudiated. The best, cautious claim seems to be: AI-driven personalization has the potential to influence purchase intent when it's correct, helpful, balanced and predictable, but could reduce confidence when it is invasive and non-transparent.

The study contributes an integrated conceptual perspective by combining technology acceptance, planned behaviour, and the personalizationprivacy paradox. Its practical message is that businesses should measure more than clicks and immediate

conversions. They should also monitor trust, complaints, opt-outs, fairness, and the quality of consumer choice. Organizations that embed responsible data governance into personalization may be better positioned to build sustainable customer relationships in the digital economy. Since the study uses secondary data only, future empirical research is needed to test the framework across industries, consumer groups, and cultural contexts.

REFERENCES

  1. Aguirre, E., Mahr, D., Grewal, D., de Ruyter, K., & Wetzels, M. (2015). Unraveling the personalization paradox: The effect of information collection and trust-building strategies on online advertisement effectiveness. Journal of Retailing, 91(1), 3449. https://doi.org/10.1016/j.jretai.2014.09.005

  2. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179211. https://doi.org/10.1016/0749-5978(91)90020-T

  3. Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319340. https://doi.org/10.2307/249008

  4. Ford, J., Jain, V., Wadhwani, K., & Gupta, D. G. (2023). AI advertising: An overview and guidelines. Journal of Business Research, 166, Article 114124. https://doi.org/10.1016/j.jbusres.2023.114124

  5. Gao, B., Wang, Y., Xie, H., Hu, Y., & Hu, Y. (2023). Artificial intelligence in advertising: Advancements, challenges, and ethical considerations in targeting, personalization, content creation, and ad optimization. SAGE Open, 13(4). https://doi.org/10.1177/21582440231210759

  6. McKinsey & Company. (2021, November 12). The value of getting personalization rightor wrongis multiplying. https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/the-value-of-getting-personalization-right- or-wrong-is-multiplying

  7. National Institute of Standards and Technology. (2023). Artificial intelligence risk management framework (AI RMF 1.0) (NIST AI 100-1). U.S. Department of Commerce. https://doi.org/10.6028/NIST.AI.100-1

  8. Tam, K. Y., & Ho, S. Y. (2006). Understanding the impact of web personalization on user information processing and decision outcomes. MIS Quarterly, 30(4), 865890. https://doi.org/10.2307/25148735

  9. Vaid, S., Puntoni, S., & Khodr, A. (2023). Artificial intelligence and empirical consumer research: A topic modeling analysis.

    Journal of Business Research, 166, Article 114110. https://doi.org/10.1016/j.jbusres.2023.114110

  10. Wadhwani, K., & Eastman, J. K. (2024). Artificial intelligence consumer behavior: A hybrid review and research agenda.

Journal of