From Recommendations to Delegation: A Systematic Review Mapping Agentic AI in E-Commerce and Its Consumer Effects

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Non-anthropomorphic, computer-initiated assistants are more prone to reactance in consumers yet enhance levels of satisfaction due to improved performance certainty.

“Pizzi et al. [24] found in another study that non-anthropomorphic, computer-initiated assistants are more prone to reactance in consumers yet enhance levels of satisfaction due to improved performance certainty.”

• cites Pizzi et al. [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceCould not verify

  The indexed copy contains a repository listing for the article and its PDF (section “Abstract > 1.01 Articolo in rivista”), but no study text addressing the claim. Searches for reactance, anthropomorphism, satisfaction, and performance certainty found no relevant passage, so the reported finding cannot be checked against our copy. (LLM, conf=0.72)

  Experimental — accuracy not yet measured

May combine two claims.

Properly designed assistants may increase loyalty when they balance effectively between entertaining, providing information, and usability, as well as overcoming concerns about privacy risks.

“properly designed assistants may increase loyalty when they balance effectively between entertaining, providing information, and usability, as well as overcoming concerns about privacy risks”

• cites Puertas et al. [4] 2024, Purchase Intentions in a Chatbot Environment: An Examination of the Effects of Customer Experience, Oeconomia Copernic. sourceSupported
• cites Vebrianti et al. [44] Vebrianti, 2025, AI Chatbots in E-Commerce: Enhancing Customer Engagement, Satisfaction and Loyalty, PaperASIA sourcePartly supported

  The Technology Acceptance Model section (page not provided) discusses information, enjoyment, and ease of use as factors in satisfaction or continued use. The Discussion (page not provided) reports that customer satisfaction affects loyalty. The paper does not establish that balancing those features increases loyalty, and our copy does not address overcoming privacy risks as a condition for doing so. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

Communication modality between text and voice affects cognitive focus on enhanced levels regarding choice satisfaction.

“The study of Schindler et al. [8] showed that communication modality between text and voice affects cognitive focus on enhanced levels regarding choice satisfaction”

• cites Schindler et al. [8] Schindler, 2024, How Speaking versus Writing to Conversational Agents Shapes Consumers’ Choice and Choice Satisfaction, J. Acad. Mark. Sci. sourcePartly supported

  The Academic implications section says speaking and writing induce different feeling- or reason-based focuses and can increase choice satisfaction. But the increase depends on the communication modality matching the product type; the citing sentence omits that condition. (LLM, conf=0.85)

  Experimental — accuracy not yet measured

May combine two claims.

The empirical research on the effects of agentic commerce is still dispersed but suggests an interrelated set of commerce and psychological effects.

“The empirical research on the effects of agentic commerce is still dispersed but suggests an interrelated set of commerce and psychological effects”

• cites [7] Klein, 2023, The Impact of Anthropomorphism on Customer Satisfaction in Chatbot Commerce: An Experimental Study in the Food Sector, Electron. Commer. Res. sourcePartly supported

  The Introduction (page unavailable) says empirical research on anthropomorphic chatbots in food e-commerce is limited. The Discussion (page unavailable) links chatbot design to enjoyment, attitude, trust, and customer satisfaction. These passages support a related set of psychological and commerce outcomes, but they do not establish that empirical research on agentic commerce generally is dispersed. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites [44] Vebrianti, 2025, AI Chatbots in E-Commerce: Enhancing Customer Engagement, Satisfaction and Loyalty, PaperASIA sourcePartly supported

  The Introduction (page unavailable) says previous e-commerce chatbot studies examined service attributes separately, and the Discussion (page unavailable) links perceived usefulness and ease of use with satisfaction and loyalty. That supports the claim's general idea of dispersed research and related customer and commercial outcomes, but the paper studies customer-service chatbots, not the broader effects of agentic commerce. (LLM, conf=0.82)

  Experimental — accuracy not yet measured
• cites [50] Sundjaja, 2025, The Determinant Factors of Continuance Use of Customer Service Chatbot in Indonesia E-Commerce: Extended Expectation Confirmation Theory, J. Sci. Technol. Policy Manag. sourceNot checked — no full text of the source
• cites [51] Cheng, 2022, Exploring Consumers’ Response to Text-Based Chatbots in e-Commerce: The Moderating Role of Task Complexity and Chatbot Disclosure, Internet Res. sourcePartly supported

  The Introduction (p. 2) links attributes of AI-enabled systems to consumer perceptions, decisions, and purchase intentions, and says empirical evidence on chatbot pushback is limited. The Abstract (p. 1) reports relationships among empathy, friendliness, trust, reliance, and resistance in e-commerce. These passages support related psychological and consumer-response effects, but the paper does not characterize research on agentic commerce broadly as dispersed or establish a broader set of commerce effects. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

May combine two claims.

Additional evidence suggests that risk/privacy may be insignificant compared to gratification and utility drivers in specific contexts/samples, and that effects may differ based on usage frequency/experience.

“Nonetheless, additional evidence suggests that risk/privacy may be insignificant compared to gratification and utility drivers in specific contexts/samples [42], and that effects may differ based on usage frequency/experience [4].”

• cites [42] Le, 2025, The Impact of Digital Innovation on E-Commerce Young Customer Satisfaction In Vietnam, J. Cent. Bank. Law Inst. sourceSupported
• cites [4] 2024, Purchase Intentions in a Chatbot Environment: An Examination of the Effects of Customer Experience, Oeconomia Copernic. sourcePartly supported

  The Results (Table 8; page unavailable) show that entertainment has a larger effect than privacy risk in this chatbot sample, supporting the comparison with gratification. But the paper reports a statistically significant negative effect of privacy risk on customer experience (p=0.046), so it does not support calling privacy risk insignificant. The usage-frequency clause is attached to [4], not this citation. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

May combine two claims.

Studies focusing on intention usually follow extended TAM/ECT/TPB logic, in which quality/usefulness and confirmation lead to higher satisfaction and/or attitudes, which then predict continued use, purchase intention, or loyalty.

“Studies that focus on intention usually follow the extended TAM/ECT/TPB logic, which says that quality/usefulness and confirmation lead to higher satisfaction and/or attitudes, which then predict outcomes like continued use, purchase intention, or loyalty”

• cites [21] Khan, 2025, Impact of Knowledge and Electronic Word of Mouth About AI-Based Chatbots on Customers’ Intention to Continue Using Chatbots in E-Commerce, J. Organ. Comput. Electron. Commer. sourceNot checked — no full text of the source
• cites [30] Hassan, 2025, The Moderating Role of Personalized Recommendations in the Trust–Satisfaction–Loyalty Relationship: An Empirical Study of AI-Driven e-Commerce, Future Bus. J. sourcePartly supported

  The Summary of the literature review (page unavailable) links fulfilled expectations to satisfaction and satisfaction to loyalty, and discusses TAM and Expectation-Confirmation Theory. The Structural equation modelling section (page unavailable) also reports positive effects of chatbot service quality and satisfaction on loyalty. The paper does not establish the broader claim that intention studies usually follow combined TAM/ECT/TPB logic, or the asserted paths to attitudes, continued-use intention, and purchase intention. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites [44] Vebrianti, 2025, AI Chatbots in E-Commerce: Enhancing Customer Engagement, Satisfaction and Loyalty, PaperASIA sourcePartly supported

  The Technology Acceptance Model (TAM) section links perceived usefulness to satisfaction, and §4.2.1 reports that information quality and perceived usefulness significantly influence satisfaction, which in turn positively affects loyalty. The cited paper does not establish the claimed extended TAM/ECT/TPB framework or a pathway through confirmation and attitudes to purchase intention; its tested outcome is loyalty. (LLM, conf=0.87)

  Experimental — accuracy not yet measured
• cites [46] Fan, Y., and Liu, X. (2022). Exploring the Role of AI Algorithmic Agents: The Impact of Algorithmic Decision Autonomy on Consumer Purchase Decisions. Front. Psychol., 13. sourceCould not verify

  The Theoretical background section uses self-determination theory to explain algorithmic decision autonomy, while Study 2 measures decision quality and satisfaction as purchase-decision outcomes (page numbers unavailable). Searches for TAM, ECT, TPB, confirmation, usefulness, continued use, and loyalty found no passage describing the claimed chain from quality or confirmation through satisfaction or attitudes to intention or loyalty. (LLM, conf=0.74)

  Experimental — accuracy not yet measured
• cites [48] Gao, 2025, The Influence of Artificial Intelligence Chatbot Problem Solving on Customers’ Continued Usage Intention in e-Commerce Platforms: An Expectation-Confirmation Model Approach, J. Bus. Res. sourceNot checked — no full text of the source
• cites [50] Sundjaja, 2025, The Determinant Factors of Continuance Use of Customer Service Chatbot in Indonesia E-Commerce: Extended Expectation Confirmation Theory, J. Sci. Technol. Policy Manag. sourceNot checked — no full text of the source
• cites [66] Kim, 2023, Understanding the Continuance Intention to Use Chatbot Services, Asia Mark. J. sourcePartly supported

  The Assessment of structural model and Conclusion report that chatbot interaction and information quality positively affect attitudes, and that attitudes predict continuance intention. The paper uses an extended TPB model, but these passages do not establish the broader claim about studies usually following TAM/ECT/TPB logic, nor do they test confirmation, satisfaction, purchase intention, or loyalty as steps or outcomes. Page numbers were not provided in our copy. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [69] Jia, 2025, A Study on the Factors That Influence Consumers’ Continuance Intention to Use Artificial Intelligence Chatbots in a Pharmaceutical e-Commerce Context, Electron Libre sourceNot checked — no full text of the source
• cites [70] Akdemir, 2024, Business and Customer-Based Chatbot Activities: The Role of Customer Satisfaction in Online Purchase Intention and Intention to Reuse Chatbots, J. Theor. Appl. Electron. Commer. Res. sourceNot checked — no full text of the source

May combine two claims.

Design and channel factors (voice vs. text; initiation mode; anthropomorphic cues) influence perceived control and social presence, occasionally altering the direction of effects through reactance or satisfaction mechanisms.

“Fourth, design and channel factors (voice vs. text; initiation mode; anthropomorphic cues) influence perceived control and social presence, occasionally altering the direction of effects through reactance or satisfaction mechanisms [7,24,31,32,52].”

• cites [7] Klein, 2023, The Impact of Anthropomorphism on Customer Satisfaction in Chatbot Commerce: An Experimental Study in the Food Sector, Electron. Commer. Res. sourcePartly supported

  The Literature review > Anthropomorphism (page not provided) says anthropomorphic design cues precede social presence, and the Discussion (page not provided) reports higher customer satisfaction with an anthropomorphic chatbot. The paper does not establish effects of voice versus text or initiation mode, changes in perceived control, or reversals of effect through reactance. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceCould not verify

  The indexed passages are a repository entry identifying the article and linking to a PDF, not passages from the article’s body; no page-numbered findings are available. Searches for perceived control, social presence, voice versus text, initiation mode, anthropomorphic cues, reactance, and satisfaction found no passage addressing the claimed effects. (LLM, conf=0.74)

  Experimental — accuracy not yet measured
• cites [31] Ding, Y., and Najaf, M. (2024). Interactivity, Humanness, and Trust: A Psychological Approach to AI Chatbot Adoption in e-Commerce. BMC Psychol., 12. sourcePartly supported

  The Discussion (page unavailable) links interactivity to perceived control and human-like chatbot features to trust; it also reports that enjoyment strengthens the positive relationship between trust and adoption. The paper does not establish voice-versus-text or initiation-mode effects, effects on social presence, reactance, or a reversal in the direction of effects. (LLM, conf=0.87)

  Experimental — accuracy not yet measured
• cites [32] Song, 2024, Uncanny Valley Effects on Chatbot Trust, Purchase Intention, and Adoption Intention in the Context of E-Commerce: The Moderating Role of Avatar Familiarity, Int. J. Hum. Comput. Interact. sourcePartly supported

  The Literature review (§§2.1–2.3; page unavailable) discusses chatbot social presence and reports that anthropomorphic visual cues can have positive or negative effects, including uncanny-valley reactions. It does not establish effects of voice versus text or initiation mode, or explain changes in effect through perceived control, reactance, or satisfaction. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [52] Gharib, 2025, Conversational AI in E-Commerce: Strategic Implications of Voice-Based Chatbots for Consumer Engagement and Trust, Int. Rev. Manag. Mark. sourcePartly supported

  Section 2.2 suggests that voice rather than text interaction may give consumers a greater sense of control, and Section 2.3 discusses anthropomorphism in relation to perceived chatbot competence. The paper does not establish effects of initiation mode or social presence, or show effects reversing through reactance or satisfaction. Page numbers are unavailable in our copy. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

May combine two claims.

Beliefs about one's ability (such as perceived task-solving competence) strengthen the usefulness and trust routes; delegation is more likely when the agent is seen as competent for the task.

“First, beliefs about one's ability (such as perceived task-solving competence) strengthen the usefulness and trust routes stronger.Delegation is more likely when the agent is seen as competent for the task [51,52,69].”

• cites [51] Cheng, 2022, Exploring Consumers’ Response to Text-Based Chatbots in e-Commerce: The Moderating Role of Task Complexity and Chatbot Disclosure, Internet Res. sourcePartly supported

  Section 2.5 (p. 7) says consumers who trust a chatbot are more likely to rely on it to address problems or make decisions. Section 4.1 (p. 13) suggests that, for complex tasks, consumers may care more about the chatbot’s ability to solve problems than its attitude. The paper does not establish that perceived competence strengthens a usefulness route or that competence itself increases delegation; its tested chatbot attributes are empathy and friendliness. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites [52] Gharib, 2025, Conversational AI in E-Commerce: Strategic Implications of Voice-Based Chatbots for Consumer Engagement and Trust, Int. Rev. Manag. Mark. sourcePartly supported

  Section 2.3 says that perceiving a chatbot as competent at solving problems increases users’ trust and belief that it can meet their needs. The paper tests task-solving competence as a moderator of the link between chatbot mode and technological optimism, not a usefulness route. The cited passages do not establish that users’ beliefs about their own ability strengthen those routes or that competence makes delegation more likely. Page numbers were not available in our copy. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [69] Jia, 2025, A Study on the Factors That Influence Consumers’ Continuance Intention to Use Artificial Intelligence Chatbots in a Pharmaceutical e-Commerce Context, Electron Libre sourceNot checked — no full text of the source

May combine two claims.

Task complexity and disclosure influence the translation of empathy and friendliness cues into trust.

“task complexity and disclosure influence the translation of empathy and friendliness cues into trust [51]”

• cites [51] Cheng, 2022, Exploring Consumers’ Response to Text-Based Chatbots in e-Commerce: The Moderating Role of Task Complexity and Chatbot Disclosure, Internet Res. sourceSupported

May combine two claims.

Systems designed to minimize friction augment perceived utility and user satisfaction, thereby reinforcing trust and subsequent intention outcomes.

“Systems designed to minimize friction (e.g., expedited information access, simplified input, structured guidance, enhanced navigation) augment perceived utility and user satisfaction, thereby reinforcing trust and subsequent intention outcomes (e.g., retention, purchase intention, reliance-oriented metrics) [6,44,45,48,50,52,67,69].”

• cites [6] Qi, 2025, AI Chatbots with Visual Search: Impact on Luxury Fashion Shopping Behavior, J. Glob. Sch. Mark. Sci. Bridg. Asia World sourceNot checked — no full text of the source
• cites [44] Vebrianti, 2025, AI Chatbots in E-Commerce: Enhancing Customer Engagement, Satisfaction and Loyalty, PaperASIA sourcePartly supported

  Section 2.5 (page unavailable) links easy navigation and low-effort chatbot use to satisfaction, acceptance, and trust. Section 4.2.1 reports that perceived ease of use and usefulness positively influence satisfaction, which in turn positively influences loyalty. The paper does not establish the full proposed chain from friction reduction through increased perceived usefulness and trust to purchase intention or reliance-oriented outcomes. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites [45] Ranjan, N., Gupta, A., Sharma, L.K., and Kapoor, S. (2024). Adoption of AI Chatbots: Revolutionizing E-Grocery Shopping in Delhi-NCR. Proceedings of the 2024 International Conference on Artificial Intelligence and Emerging Technology, Global AI Summit 2024, Institute of Electrical and Electronics Engineers Inc. sourceNot checked — no full text of the source
• cites [48] Gao, 2025, The Influence of Artificial Intelligence Chatbot Problem Solving on Customers’ Continued Usage Intention in e-Commerce Platforms: An Expectation-Confirmation Model Approach, J. Bus. Res. sourceNot checked — no full text of the source
• cites [50] Sundjaja, 2025, The Determinant Factors of Continuance Use of Customer Service Chatbot in Indonesia E-Commerce: Extended Expectation Confirmation Theory, J. Sci. Technol. Policy Manag. sourceNot checked — no full text of the source
• cites [52] Gharib, 2025, Conversational AI in E-Commerce: Strategic Implications of Voice-Based Chatbots for Consumer Engagement and Trust, Int. Rev. Manag. Mark. sourcePartly supported

  Section 5.2 (page not provided) recommends quick, easy chatbot interactions and says efficient problem-solving can strengthen trust, satisfaction, and continuance intentions. Section 5.1 also reports a pathway from chatbot interaction through trust and perceived customer care to continuance intention. The paper does not establish the full claimed pathway through perceived utility and user satisfaction, or the broader purchase-intention and reliance outcomes. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites [67] Ionescu, I.-M., and Ardelean, A. (2026, February 23). The Romanian Consumer’s Perspective on the Integration of Artifi Cial Intelligence in E-Commerce. Rom. Stat. Rev., Available online: https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://www.revistadestatistica.ro/2024/12/the-romanian-consumers-perspective-on-the-integration-of-artificial-intelligence-in-e-commerce/&ved=2ahUKEwigo7jzyviSAxXfKtAFHXflHOMQFnoECBwQAQ&usg=AOvVaw2T7upPRWaNo-21PkBVTiff.Not checked yet

  Not checked in the free run.

• cites [69] Jia, 2025, A Study on the Factors That Influence Consumers’ Continuance Intention to Use Artificial Intelligence Chatbots in a Pharmaceutical e-Commerce Context, Electron Libre sourceNot checked — no full text of the source

May combine two claims.

Numerous models conceptualize risk and privacy as impediments that amplify adverse emotions and resistance or directly inhibit purchase or adoption intentions.

“Numerous models conceptualize risk and privacy as impediments that amplify adverse emotions and resistance or directly inhibit purchase or adoption intentions [4,49,67].”

• cites [4] 2024, Purchase Intentions in a Chatbot Environment: An Examination of the Effects of Customer Experience, Oeconomia Copernic. sourceSupported
• cites [49] Chang, 2025, Understand Resist Use Online Customer Service Chatbot: An Integrated Innovation Resist Theory and Negative Emotion Perspective, Aslib J. Inf. Manag. sourceNot checked — no full text of the source
• cites [67] Ionescu, I.-M., and Ardelean, A. (2026, February 23). The Romanian Consumer’s Perspective on the Integration of Artifi Cial Intelligence in E-Commerce. Rom. Stat. Rev., Available online: https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://www.revistadestatistica.ro/2024/12/the-romanian-consumers-perspective-on-the-integration-of-artificial-intelligence-in-e-commerce/&ved=2ahUKEwigo7jzyviSAxXfKtAFHXflHOMQFnoECBwQAQ&usg=AOvVaw2T7upPRWaNo-21PkBVTiff.Not checked yet

  Not checked in the free run.

May combine two claims.

Response time, quality of information, usefulness, and ease of use positively influence satisfaction and loyalty, with customer experience mediating.

“Vebrianti et al. [44] has shown that the combined effect of the response time, the quality of information, usefulness, and ease of use has positively influenced the satisfaction level and the level of loyalty, while the customer experience has acted as the mediator”

• cites Vebrianti et al. [44] Vebrianti, 2025, AI Chatbots in E-Commerce: Enhancing Customer Engagement, Satisfaction and Loyalty, PaperASIA sourcePartly supported

  The Results (§4.2.1) report that response time, information quality, perceived usefulness, and perceived ease of use positively influence customer satisfaction, which in turn positively affects loyalty. But the research framework (H5) identifies customer satisfaction—not customer experience—as the mediating variable. (LLM, conf=0.90)

  Experimental — accuracy not yet measured

May combine two claims.

Social presence and entertainment are powerful antecedents of positive customer experience and purchase intention.

“Puertas et al. [4] found social presence and entertainment as powerful antecedents of positive customer experience and purchase intention”

• cites Puertas et al. [4] 2024, Purchase Intentions in a Chatbot Environment: An Examination of the Effects of Customer Experience, Oeconomia Copernic. sourcePartly supported

  The Results (page unavailable) find that entertainment improves customer experience and has a positive indirect effect on purchase intention. Social presence improves customer experience and indirectly encourages purchase intention only among infrequent or novice chatbot users; its effect is not significant among frequent users. The claim omits that qualification and does not distinguish indirect effects on purchase intention from direct effects on experience. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

May combine two claims.

True agentic execution (placing an order and paying for it) is still rare and mostly limited to scenario manipulations rather than embedded transaction observation.

“In the same way, true agentic execution (placing an order and paying for it) is still rare and mostly limited to scenario manipulations rather than embedded transaction observation [40,41].”

• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source
• cites [41] Frank, 2025, Consumer Acceptance of High-Autonomy AI Assistants Is Driven by Perceived Benefits in Online Shopping Settings Characterized by Scarcity, Psychol. Mark. sourcePartly supported

  The Methods for Studies 2A and 2B describe imagined shopping scenarios in which an AI could make purchases, not observed orders or payments. Study 1 measured ad clicks, and Limitations and Future Research (section 11.3; page unavailable) calls for future studies that track actual purchase behavior. This supports the distinction between simulated execution and observed transactions in this paper, but does not establish that observed agentic transactions are rare across the wider literature. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

May combine two claims.

In the main empirical cluster (Autonomy-1, recommendation/support scope), positive outcomes usually follow a utility/assurance chain: performance cues and interaction quality make people feel more useful and satisfied, which builds trust and leads to downstream adoption/continuance or purchase-intent outcomes.

“In the main empirical cluster (Autonomy-1, recommendation/support scope), positive outcomes usually follow a utility/assurance chain: performance cues and interaction quality make people feel more useful and satisfied, which builds trust and leads to downstream adoption/continuance or purchase-intent outcomes (e.g., [44,46,48,50,69]).”

• cites [44] Vebrianti, 2025, AI Chatbots in E-Commerce: Enhancing Customer Engagement, Satisfaction and Loyalty, PaperASIA sourcePartly supported

  Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured
• cites [46] Fan, Y., and Liu, X. (2022). Exploring the Role of AI Algorithmic Agents: The Impact of Algorithmic Decision Autonomy on Consumer Purchase Decisions. Front. Psychol., 13. sourcePartly supported

  Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured
• cites [48] Gao, 2025, The Influence of Artificial Intelligence Chatbot Problem Solving on Customers’ Continued Usage Intention in e-Commerce Platforms: An Expectation-Confirmation Model Approach, J. Bus. Res. sourceNot checked — no full text of the source
• cites [50] Sundjaja, 2025, The Determinant Factors of Continuance Use of Customer Service Chatbot in Indonesia E-Commerce: Extended Expectation Confirmation Theory, J. Sci. Technol. Policy Manag. sourceNot checked — no full text of the source
• cites [69] Jia, 2025, A Study on the Factors That Influence Consumers’ Continuance Intention to Use Artificial Intelligence Chatbots in a Pharmaceutical e-Commerce Context, Electron Libre sourceNot checked — no full text of the source

May combine two claims.

In role-based autonomy frameworks, the co-assistant role can attain the optimal equilibrium between support and perceived agency, corroborating that delegation is most effective when users maintain significant authorship and control rights.

“In role-based autonomy frameworks (e.g., substitute vs. co-assistant vs. performer), the co-assistant role can attain the optimal equilibrium between support and perceived agency, corroborating the overarching conclusion that delegation is most effective when users maintain significant authorship and control rights [46].”

• cites [46] Fan, Y., and Liu, X. (2022). Exploring the Role of AI Algorithmic Agents: The Impact of Algorithmic Decision Autonomy on Consumer Purchase Decisions. Front. Psychol., 13. sourceSupported

May combine two claims.

Design can also boost perceived autonomy and psychological ownership, which can boost trust and sometimes attitudes toward the product.

“Other research, however, shows that design can also boost perceived autonomy and psychological ownership, which can boost trust and sometimes attitudes toward the product [38].”

• cites [38] Hu, 2025, Empowering Consumers: The Role of Avatar Choice in Consumer-Chatbot Interaction and Psychological Ownership, Int. J. Hum. Comput. Interact. sourcePartly supported

  The Results (section 4.2; page unavailable) support an effect of chatbot-avatar choice on psychological ownership and perceived autonomy, and a positive indirect effect on product attitude through ownership and trust. They do not establish that autonomy itself boosts trust or product attitude; the autonomy-mediated effect on product attitude was not significant. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

May combine two claims.

Behavioral outcomes such as conversion, click-through, repurchase, and GMV are less common but provide stronger external validity.

“Behavioral outcomes (like conversion, click-through, repurchase, and GMV) are less common but provide stronger external validity (e.g., [10,41,55]).”

• cites [10] Nie, G., Zhi, R., Yan, X., Du, Y., Zhang, X., Chen, J., Zhou, M., Chen, H., Li, T., and Cheng, Z. (2024). A Hybrid Multi-Agent Conversational Recommender System with LLM and Search Engine in E-Commerce. Proceedings of the RecSys 2024—Proceedings of the 18th ACM Conference on Recommender Systems, Association for Computing Machinery, Inc. sourceNot checked — no full text of the source
• cites [41] Frank, 2025, Consumer Acceptance of High-Autonomy AI Assistants Is Driven by Perceived Benefits in Online Shopping Settings Characterized by Scarcity, Psychol. Mark. sourcePartly supported

  The Results (section 6.2; page unavailable) use click-through rate as a behavioral outcome, and the Method (section 6.1; page unavailable) describes the ad campaign as a realistic test prioritizing ecological validity. The paper does not establish that behavioral outcomes are less common, or address conversion, repurchase, or GMV as outcomes; it also cautions in Limitations (section 11.3; page unavailable) that click-through rates may not capture long-term behavior. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [55] Ovalle, C. (2024). Integrated Recommendation System in ChatGPT to Analyze Post-Purchase Behavior of E-Commerce Store Users. Proceedings of the LACCEI International Multi-Conference for Engineering, Education and Technology, Latin American and Caribbean Consortium of Engineering Institutions. sourcePartly supported

  The E. Product Completion section reports repurchase rates and purchases of recommended products, supporting the use of purchase behavior as an outcome. It does not say that behavioral outcomes are less common or provide stronger external validity, and the retrieved passages do not address click-through or GMV. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

May combine two claims.

Co-assistant designs can maintain satisfaction and choice certainty when users recognize performance advantages without a reduction in governance, and modality/interaction selections can influence whether autonomy is perceived as supportive or intrusive.

“Conversely, co-assistant designs can maintain satisfaction and choice certainty when users recognize performance advantages without a reduction in governance [24], and modality/interaction selections can additionally influence whether autonomy is perceived as supportive or intrusive [8].”

• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceCould not verify

  Split into 2 atoms — 0 confirmed, 0 partially supported, 0 disputed, 0 misattributed, 2 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured
• cites [8] Schindler, 2024, How Speaking versus Writing to Conversational Agents Shapes Consumers’ Choice and Choice Satisfaction, J. Acad. Mark. Sci. sourcePartly supported

  Split into 2 atoms — 0 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 1 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured

May combine two claims.

Design elements that suggest system-initiated control, particularly in the absence of explicit user veto or approval, exacerbate this phenomenon, even when downstream satisfaction may remain favorable due to certainty and performance improvements.

“Design elements that suggest system-initiated control, particularly in the absence of explicit user veto or approval, exacerbate this phenomenon, even when downstream satisfaction may remain favorable due to certainty and performance improvements [24].”

• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceCould not verify

  Split into 2 atoms — 0 confirmed, 0 partially supported, 0 disputed, 0 misattributed, 2 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured

May combine two claims.

Negative reactions arise when autonomy cues suggest a loss of control (reactance/threat-to-freedom) or displacement (powerlessness); conversely, positive outcomes are observed when systems facilitate friction reduction while maintaining perceived authorship and recourse.

“Negative reactions arise when autonomy cues suggest a loss of control (reactance/threat-to-freedom) or displacement (powerlessness); conversely, positive outcomes are observed when systems facilitate friction reduction while maintaining perceived authorship and recourse [23,24,40].”

• cites [23] Yoon, 2025, Avoidance of AI-Empowered Digital Service Assistants in Fashion Shopping: The Negative Side of Personalized Recommendation of Chatbots, J. Glob. Fash. Mark. sourceNot checked — no full text of the source
• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceCould not verify

  Split into 2 atoms — 0 confirmed, 0 partially supported, 0 disputed, 0 misattributed, 2 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured
• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source

May combine two claims.

Autonomy effects are contingent upon context especially regarding urgency and scarcity, and may not extend to routine shopping without education, safeguards, and trust recalibration.

“The mapped evidence suggests that autonomy effects are contingent upon contextespecially regarding urgency and scarcity-and may not extend to routine shopping without education, safeguards, and trust recalibration [40,41].”

• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source
• cites [41] Frank, 2025, Consumer Acceptance of High-Autonomy AI Assistants Is Driven by Perceived Benefits in Online Shopping Settings Characterized by Scarcity, Psychol. Mark. sourcePartly supported

  The Theoretical Contribution (§11.1) says consumers are more willing to delegate decisions to high-autonomy AI under scarcity, while the Practical Implications (§11.2) say such assistants can be expected to underperform in routine shopping. The paper discusses transparency safeguards, but it does not establish that education, safeguards, and trust recalibration are needed for autonomy effects to extend to routine shopping. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

May combine two claims.

A significant body of literature examines anthropomorphism, modality, warmth/competence cues, and initiation patterns, frequently demonstrating impacts on satisfaction, social presence, or reactance in controlled environments.

“A significant body of literature examines anthropomorphism, modality, warmth/competence cues, and initiation patterns, frequently demonstrating impacts on satisfaction, social presence, or reactance in controlled environments [8,24].”

• cites [8] Schindler, 2024, How Speaking versus Writing to Conversational Agents Shapes Consumers’ Choice and Choice Satisfaction, J. Acad. Mark. Sci. sourcePartly supported

  The Academic implications section (page unavailable) reports that matching speaking or writing with product type increases choice satisfaction, and the Empirical overview describes five studies of communication modality. The paper does not establish the sentence’s other attributed topics—anthropomorphism, warmth/competence cues, or initiation patterns—or effects on social presence or reactance. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourcePartly supported

  The indexed Abstract (page unavailable) reports two studies in which assistants’ anthropomorphic appearance and initiation pattern affected psychological reactance, with downstream effects on satisfaction. It does not address modality, warmth/competence cues, or social presence, so it supports only part of the broad statement attributed to this source. (LLM, conf=0.85)

  Experimental — accuracy not yet measured

May combine two claims.

Panels are often not random samples, and the range of topics they cover can be very limited, such as one product category, one platform, or a few task types.

“Panels are often not random samples, and the range of topics they cover can be very limited (for example, just one product category, one platform, or a few types of tasks) [7,8,41].”

• cites [7] Klein, 2023, The Impact of Anthropomorphism on Customer Satisfaction in Chatbot Commerce: An Experimental Study in the Food Sector, Electron. Commer. Res. sourcePartly supported

  The Method (Data collection and sample) describes non-probability convenience and snowball sampling. The Discussion (Limitations and future research) says the experiment was limited to food products and used a restricted e-store scenario. These passages support this study as an example of a non-random sample and narrow scope, but do not establish that panels are often like this. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [8] Schindler, 2024, How Speaking versus Writing to Conversational Agents Shapes Consumers’ Choice and Choice Satisfaction, J. Acad. Mark. Sci. sourcePartly supported

  The Method section describes a Prolific sample and a tablet-choice task, and Limitations and future research explicitly says Study 5 was restricted to a hedonic product context. These passages support an example of a narrowly scoped study, but the paper does not establish that panels are often nonrandom or generally cover only a limited range of topics; its studies use several choice contexts. Page numbers are unavailable in our copy. (LLM, conf=0.82)

  Experimental — accuracy not yet measured
• cites [41] Frank, 2025, Consumer Acceptance of High-Autonomy AI Assistants Is Driven by Perceived Benefits in Online Shopping Settings Characterized by Scarcity, Psychol. Mark. sourcePartly supported

  The Limitations and Future Research section (11.3; page unavailable) explicitly says the studies were confined to online shopping and a narrow set of utilitarian electronic products, supporting the limited-topic example. The Method section (7.1; page unavailable) says participants were recruited through Prolific, but the copy does not establish that panels are often nonrandom samples. It also does not establish a general pattern of studies limited to one platform or a few task types. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

May combine two claims.

Perceived usefulness/satisfaction mediating design cues to trust/continuance and trust mediating performance cues to intention are mediation structures frequently reported in the mapped studies.

“Perceived usefulness/satisfaction mediating design cues → trust/continuance and trust mediating performance cues → intention are two examples of mediation structures frequently reported in the mapped studies [44,46,48,50,69].”

• cites [44] Vebrianti, 2025, AI Chatbots in E-Commerce: Enhancing Customer Engagement, Satisfaction and Loyalty, PaperASIA sourcePartly supported

  The H5 research-framework passage (page unavailable) identifies customer satisfaction as mediating the effects of response time, information quality, perceived usefulness, and perceived ease of use on customer loyalty; the Discussion reports the corresponding relationships. The paper does not show satisfaction mediating effects on trust or continuance, or trust mediating performance cues to intention. It supports a related mediation structure, but not the structures as stated. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites [46] Fan, Y., and Liu, X. (2022). Exploring the Role of AI Algorithmic Agents: The Impact of Algorithmic Decision Autonomy on Consumer Purchase Decisions. Front. Psychol., 13. sourceCould not verify

  Study 2, “Mediating effect of self-efficacy” (page unavailable), reports a different pathway: algorithmic decision autonomy affects purchase decisions partly through self-efficacy. Searches for the claimed usefulness/satisfaction and trust mediation structures found no passage establishing either structure or how frequently it appears in mapped studies. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [48] Gao, 2025, The Influence of Artificial Intelligence Chatbot Problem Solving on Customers’ Continued Usage Intention in e-Commerce Platforms: An Expectation-Confirmation Model Approach, J. Bus. Res. sourceNot checked — no full text of the source
• cites [50] Sundjaja, 2025, The Determinant Factors of Continuance Use of Customer Service Chatbot in Indonesia E-Commerce: Extended Expectation Confirmation Theory, J. Sci. Technol. Policy Manag. sourceNot checked — no full text of the source
• cites [69] Jia, 2025, A Study on the Factors That Influence Consumers’ Continuance Intention to Use Artificial Intelligence Chatbots in a Pharmaceutical e-Commerce Context, Electron Libre sourceNot checked — no full text of the source

May combine two claims.

Quasi-field studies enhance ecological validity but encounter challenges including ambiguous randomization, limited site contexts, and short follow-up intervals.

“Quasi-field studies enhance ecological validity but encounter challenges including ambiguous randomization, limited site contexts, and short follow-up intervals [55].”

• cites [55] Ovalle, C. (2024). Integrated Recommendation System in ChatGPT to Analyze Post-Purchase Behavior of E-Commerce Store Users. Proceedings of the LACCEI International Multi-Conference for Engineering, Education and Technology, Latin American and Caribbean Consortium of Engineering Institutions. sourcePartly supported

  The paper describes a quasi-experimental study of an e-commerce store in Lima with control and experimental groups (E. Product Completion). It reports a 14-day repurchase measure, supporting the claim’s limited-site and short-interval details as descriptions of this study. It does not discuss ecological validity, say randomization was ambiguous, or identify these features as general challenges of quasi-field studies. Page numbers were not supplied in our copy. (LLM, conf=0.78)

  Experimental — accuracy not yet measured

May combine two claims.

Numerous studies exhibit robust internal validity while functioning within simulated purchasing contexts.

“Numerous studies exhibit robust internal validity while functioning within simulated purchasing contexts [7,8,24,63].”

• cites [7] Klein, 2023, The Impact of Anthropomorphism on Customer Satisfaction in Chatbot Commerce: An Experimental Study in the Food Sector, Electron. Commer. Res. sourcePartly supported

  The Method (Data collection and sample; page unavailable) says participants were randomly assigned to two chatbot scenarios. The Discussion (Limitations and future research; page unavailable) describes an experimental e-store with no full checkout, supporting the simulated-purchasing part. The paper does not establish robust internal validity; it notes that the scenario was recognizable as an experiment and may have biased respondents. (LLM, conf=0.83)

  Experimental — accuracy not yet measured
• cites [8] Schindler, 2024, How Speaking versus Writing to Conversational Agents Shapes Consumers’ Choice and Choice Satisfaction, J. Acad. Mark. Sci. sourcePartly supported

  The Empirical overview describes five studies, and Limitations and future research identifies them as lab and online experiments designed to mirror real-world interactions rather than field studies (page not provided). The paper describes controlled conditions and calls its behavioral outcomes strong evidence, but it does not assess the internal validity of numerous studies generally or explicitly establish that its own internal validity is robust. (LLM, conf=0.82)

  Experimental — accuracy not yet measured
• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceCould not verify

  The available indexed passages are repository metadata, a PDF listing, and an unrelated ASN simulation notice; none describes the paper’s study design or purchasing context. Searches for internal validity, simulated purchasing, and shopping scenarios found no relevant passage, so this copy cannot verify the claim. (LLM, conf=0.74)

  Experimental — accuracy not yet measured
• cites [63] Matosas-López, L. (2025). The Effect of Chatbots Humanness and Brand Associations on Consumer Purchase Intentions: An Experimental Approach. Electron. Commer. Res. sourceNot checked — no full text of the source

May combine two claims.

The autonomy-to-powerlessness mechanism is well-supported in controlled experiments but often not tested in real-life transactions.

“The autonomy-to-powerlessness mechanism is well-supported in controlled experiments, but it is often not tested in real-life transactions [40,46].”

• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source
• cites [46] Fan, Y., and Liu, X. (2022). Exploring the Role of AI Algorithmic Agents: The Impact of Algorithmic Decision Autonomy on Consumer Purchase Decisions. Front. Psychol., 13. sourcePartly supported

  Study 3 (Overview of studies; no page given) reports that high algorithmic decision autonomy reduced self-determination and self-efficacy among participants with high power distance. The study used controlled scenarios, including a simulated e-commerce shopping guide, rather than observed transactions. But the paper does not establish that this mechanism is *often* untested in real-life transactions, and it measures self-efficacy rather than powerlessness directly. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

May combine two claims.

The review includes a small number of field or quasi-field studies that show the strongest link to real outcomes, but these studies are often hard to replicate across platforms and agent architectures because they fail to report enough on assignment, exposure windows, and guardrails.

“The review includes a small number of field or quasi-field studies that show the strongest link to real outcomes (like acceptance/conversion, spend/GMV), but these studies are often hard to replicate across platforms and agent architectures because they fail to report enough on assignment, exposure windows, and guardrails [10,55].”

• cites [10] Nie, G., Zhi, R., Yan, X., Du, Y., Zhang, X., Chen, J., Zhou, M., Chen, H., Li, T., and Cheng, Z. (2024). A Hybrid Multi-Agent Conversational Recommender System with LLM and Search Engine in E-Commerce. Proceedings of the RecSys 2024—Proceedings of the 18th ACM Conference on Recommender Systems, Association for Computing Machinery, Inc. sourceNot checked — no full text of the source
• cites [55] Ovalle, C. (2024). Integrated Recommendation System in ChatGPT to Analyze Post-Purchase Behavior of E-Commerce Store Users. Proceedings of the LACCEI International Multi-Conference for Engineering, Education and Technology, Latin American and Caribbean Consortium of Engineering Institutions. sourcePartly supported

  Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured

May combine two claims.

Interface design is not superficial appearance; cues can shift agency perceptions and reactance, but the market importance of these cue pathways is still not well understood.

“This bridging logic follows the review's broader conclusion that interface design is not superficial appearance; cues can shift agency perceptions and reactance-but the market importance of these cue pathways is still not well understood [23,24].”

• cites [23] Yoon, 2025, Avoidance of AI-Empowered Digital Service Assistants in Fashion Shopping: The Negative Side of Personalized Recommendation of Chatbots, J. Glob. Fash. Mark. sourceNot checked — no full text of the source
• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceCould not verify

  The indexed passages are a repository listing for the paper and a PDF link (Abstract listing; no page number), not substantive passages from the paper. Searches for interface cues, agency, reactance, and market importance returned no passage addressing the claim, so our copy cannot establish whether the source supports it. (LLM, conf=0.74)

  Experimental — accuracy not yet measured

May combine two claims.

Experimental research frequently substantiates causal assertions when randomization is evident and manipulation checks are documented, but limitations endure, including scenario-based autonomy, restricted domains, and intention-laden outcomes.

“Table 11 demonstrates that experimental research frequently substantiates causal assertions when randomization is evident and manipulation checks are documented; nonetheless, limitations endure, characterized by scenario-based autonomy, restricted domains, and intention-laden outcomes [40,41,46].”

• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source
• cites [41] Frank, 2025, Consumer Acceptance of High-Autonomy AI Assistants Is Driven by Perceived Benefits in Online Shopping Settings Characterized by Scarcity, Psychol. Mark. sourcePartly supported

  Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured
• cites [46] Fan, Y., and Liu, X. (2022). Exploring the Role of AI Algorithmic Agents: The Impact of Algorithmic Decision Autonomy on Consumer Purchase Decisions. Front. Psychol., 13. sourcePartly supported

  Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured

May combine two claims.

Acceptance models usually assume that delegation is the action-level construct, and autonomy is the boundary condition that makes governance costs important.

“Acceptance models usually assume that delegation is the action-level construct, and autonomy is the boundary condition that makes governance costs important [23,24,40].”

• cites [23] Yoon, 2025, Avoidance of AI-Empowered Digital Service Assistants in Fashion Shopping: The Negative Side of Personalized Recommendation of Chatbots, J. Glob. Fash. Mark. sourceNot checked — no full text of the source
• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceCould not verify

  The indexed copy contains a repository record and a link to the manuscript PDF (section ‘Abstract > 1.01 Articolo in rivista’, no page), but not the paper’s substantive text. Searches for delegation, autonomy, acceptance models, and governance costs returned no passage addressing the claim, so the attribution cannot be checked from this copy. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source

May combine two claims.

Experiments confirm autonomy to powerlessness/reactance mechanisms and interface-driven reactance shifts, whereas surveys focus on usefulness/satisfaction/trust chains.

“This disparity is evident in both experimental and survey methodologies: experiments confirm autonomy → powerlessness/reactance mechanisms [23,40] and interface-driven reactance shifts [24], whereas surveys focus on usefulness/satisfaction/trust chains [42,44,69].”

• cites [23] Yoon, 2025, Avoidance of AI-Empowered Digital Service Assistants in Fashion Shopping: The Negative Side of Personalized Recommendation of Chatbots, J. Glob. Fash. Mark. sourceNot checked — no full text of the source
• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source
• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceSupported
• cites [42] Le, 2025, The Impact of Digital Innovation on E-Commerce Young Customer Satisfaction In Vietnam, J. Cent. Bank. Law Inst. sourcePartly supported

  The Data Analysis Methods section (page unavailable) describes a survey, and the Structural Model Assessment reports a positive association between utilitarian gratification and user satisfaction. The paper does not test trust as part of a usefulness–satisfaction–trust chain; its model lists satisfaction as the outcome and contains no trust variable. The experimental claims are attached to other citations, not this source. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [44] Vebrianti, 2025, AI Chatbots in E-Commerce: Enhancing Customer Engagement, Satisfaction and Loyalty, PaperASIA sourcePartly supported

  Section 3.2 (page unavailable) describes an online questionnaire survey, and section 4.2.1 reports relationships between perceived usefulness, customer satisfaction, and loyalty. The paper does not test trust as part of that chain; its stated model in the H5 section ends with satisfaction and loyalty. The experimental autonomy/reactance claims are attached to other citations, not this source. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [69] Jia, 2025, A Study on the Factors That Influence Consumers’ Continuance Intention to Use Artificial Intelligence Chatbots in a Pharmaceutical e-Commerce Context, Electron Libre sourceNot checked — no full text of the source

May combine two claims.

Time-lag designs are infrequent and typically confined to intention-based outcomes rather than actual behavior.

“Time-lag designs are infrequent and typically confined to intention-based outcomes rather than actual behavior [31].”

• cites [31] Ding, Y., and Najaf, M. (2024). Interactivity, Humanness, and Trust: A Psychological Approach to AI Chatbot Adoption in e-Commerce. BMC Psychol., 12. sourceCould not verify

  The Data analysis sections report intention to adopt chatbots as this study’s outcome, but they do not establish how common time-lag designs are or what outcomes such designs typically measure. Searches for time-lagged and longitudinal designs and actual behavior found no non-abstract passage addressing that broader claim; the Method section describes an online survey conducted over four weeks. (LLM, conf=0.71)

  Experimental — accuracy not yet measured

May combine two claims.

A lot of the strongest causal evidence in the corpus comes from experiments instead of natural settings, leaving the real-world persistence of effects under-identified.

“This suggestion fits with the fact that a lot of the strongest causal evidence in the corpus comes from experiments instead of natural settings, leaving the real-world persistence of effects under-identified [8,40].”

• cites [8] Schindler, 2024, How Speaking versus Writing to Conversational Agents Shapes Consumers’ Choice and Choice Satisfaction, J. Acad. Mark. Sci. sourcePartly supported

  The Limitations and future research section says the main studies were lab and online experiments and calls for field studies to test whether their findings hold in more natural settings. It also reports an initial offline field study, so the work is not exclusively experimental. The paper does not assess the corpus-wide strength of causal evidence or the persistence of effects over time. (LLM, conf=0.83)

  Experimental — accuracy not yet measured
• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source

May combine two claims.

Field and quasi-field contributions enhance inferences regarding market outcomes, but methodological limitations constrain generalizability and the testing of mechanisms.

“Field and quasi-field contributions enhance inferences regarding market outcomes; however, methodological limitations-including ambiguous assignment, platform specificity, abbreviated observation periods, and analytical simplifications-constrain generalizability and the testing of mechanisms [10,55].”

• cites [10] Nie, G., Zhi, R., Yan, X., Du, Y., Zhang, X., Chen, J., Zhou, M., Chen, H., Li, T., and Cheng, Z. (2024). A Hybrid Multi-Agent Conversational Recommender System with LLM and Search Engine in E-Commerce. Proceedings of the RecSys 2024—Proceedings of the 18th ACM Conference on Recommender Systems, Association for Computing Machinery, Inc. sourceNot checked — no full text of the source
• cites [55] Ovalle, C. (2024). Integrated Recommendation System in ChatGPT to Analyze Post-Purchase Behavior of E-Commerce Store Users. Proceedings of the LACCEI International Multi-Conference for Engineering, Education and Technology, Latin American and Caribbean Consortium of Engineering Institutions. sourcePartly supported

  In “E. Product Completion” (page not provided), the paper describes a quasi-experimental comparison of control and experimental groups and reports e-commerce repurchase outcomes, including purchases within 14 days. It therefore supports the field-like market-outcomes portion of the claim. It does not itself identify ambiguous assignment, platform specificity, abbreviated observation, or analytical simplifications as limitations constraining generalizability or tests of mechanisms. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

May combine two claims.

Purchase intention/adoption intention is mostly positive via utility/trust, but can decrease under high autonomy.

“Mostly positive via utility/trust, but can decrease under high autonomy”

• cites [4] 2024, Purchase Intentions in a Chatbot Environment: An Examination of the Effects of Customer Experience, Oeconomia Copernic. sourcePartly supported

  The paper’s literature review (Media Appeal; Limitations and directions for future research, page unavailable) reports positive links from usefulness and trust to purchase or use intentions. Its Results also report positive indirect effects on purchase intention, but the negative effects it identifies concern privacy risk and, for frequent users, information—not high autonomy. A search for autonomy found no passage supporting that condition. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites [6] Qi, 2025, AI Chatbots with Visual Search: Impact on Luxury Fashion Shopping Behavior, J. Glob. Sch. Mark. Sci. Bridg. Asia World sourceNot checked — no full text of the source
• cites [8] Schindler, 2024, How Speaking versus Writing to Conversational Agents Shapes Consumers’ Choice and Choice Satisfaction, J. Acad. Mark. Sci. sourceCould not verify

  The Method and Study 5 passages measure intention to choose a product, but the reported effects concern speaking versus writing and brand equity. Searches for utility, trust, autonomy, and adoption intention found no passage in our copy supporting the claimed pathway or a decrease under high autonomy. (LLM, conf=0.72)

  Experimental — accuracy not yet measured
• cites [22] Yatawara, K.M.B., Sampath, T., Kalupahana, P.L., Rathnayake, S.O., Jayasuriya, N., and Rathnayake, N. (2025). Bridging the Chatbot Connection: The Role of AI-Driven Chatbot Affordances in e-Commerce Purchase Intentions. Int. J. Sociol. Soc. Policy, 1–21. sourceNot checked — no full text of the source
• cites [30] Hassan, 2025, The Moderating Role of Personalized Recommendations in the Trust–Satisfaction–Loyalty Relationship: An Empirical Study of AI-Driven e-Commerce, Future Bus. J. sourcePartly supported

  The Summary of the literature review (page unavailable) links useful AI chatbots and trust to customer loyalty, while Questionnaire design and scale description (page unavailable) says the loyalty measure includes intentions for repeat purchases. The paper does not establish that purchase or adoption intention decreases under high autonomy; its discussion of user control concerns improved trust and loyalty, not that proposed decrease. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [31] Ding, Y., and Najaf, M. (2024). Interactivity, Humanness, and Trust: A Psychological Approach to AI Chatbot Adoption in e-Commerce. BMC Psychol., 12. sourcePartly supported

  The Discussion (page unavailable) reports that trust significantly mediates intention to adopt AI chatbots for e-commerce, and the Hypothesis development section discusses utilitarian factors and perceived usefulness. Neither those passages nor searches for autonomy support the claim that intention decreases under high autonomy; the paper instead examines perceived enjoyment as a moderator. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites [32] Song, 2024, Uncanny Valley Effects on Chatbot Trust, Purchase Intention, and Adoption Intention in the Context of E-Commerce: The Moderating Role of Avatar Familiarity, Int. J. Hum. Comput. Interact. sourcePartly supported

  The Results (section 5.2; page unavailable) report that higher trust was associated with higher purchase intention and willingness to use the chatbot in the future. The paper also reports decreases linked to avatar-induced eeriness through trust, but the cited passages do not support utility as the pathway or high autonomy as the condition for a decrease. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [39] Liu, J., and Chen, J. (2025). Chatbot-Aided Product Purchases among Generation Z: The Role of Personality Traits. Front. Psychol., 16. sourcePartly supported

  The Results (structural equation modeling; page not provided) report positive associations between purchase intention and chatbot expertise, interactivity, and customization, supporting the functional-benefits portion of the claim. But trustworthiness had no statistically significant effect, and the passages do not report a decrease in intention under high autonomy. (LLM, conf=0.87)

  Experimental — accuracy not yet measured
• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source
• cites [41] Frank, 2025, Consumer Acceptance of High-Autonomy AI Assistants Is Driven by Perceived Benefits in Online Shopping Settings Characterized by Scarcity, Psychol. Mark. sourcePartly supported

  The Main Analyses (§9.2.2; page unavailable) find a positive indirect effect on AI adoption intentions through perceived benefits under scarcity, but not under regular shopping conditions. The Abstract (page unavailable) reports lower willingness to use a high-autonomy assistant than a low-autonomy one. The paper does not establish trust as the pathway, and its context-dependent findings do not support calling the effect mostly positive. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

May combine two claims.

Satisfaction declines when designs trigger reactance or cognitive/affective overload, which depress downstream evaluations.

“It declines when designs trigger reactance or cognitive/affective overload, which depress downstream evaluations.”

• cites [4] 2024, Purchase Intentions in a Chatbot Environment: An Examination of the Effects of Customer Experience, Oeconomia Copernic. sourcePartly supported

  The Discussion (page unavailable) says information overload may impair experienced chatbot users’ customer experience, and reports that information may reduce their purchase intention. The paper also summarizes evidence that privacy risk lowers satisfaction, which in turn affects continued use and loyalty (Additional information, page unavailable). It does not establish that reactance or cognitive/affective overload lowers satisfaction or thereby depresses downstream evaluations. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites [7] Klein, 2023, The Impact of Anthropomorphism on Customer Satisfaction in Chatbot Commerce: An Experimental Study in the Food Sector, Electron. Commer. Res. sourcePartly supported

  The Literature review > Chatbot commerce (page unavailable) says highly human-like agents can raise expectations and cause dissatisfaction when their capabilities disappoint. That supports a possible design-related decline in satisfaction, but the paper does not attribute it to reactance or cognitive/affective overload, or show that those mechanisms depress downstream evaluations. The Literature review > Anthropomorphism instead says anthropomorphic design can help consumers cope with information overload. (LLM, conf=0.75)

  Experimental — accuracy not yet measured
• cites [22] Yatawara, K.M.B., Sampath, T., Kalupahana, P.L., Rathnayake, S.O., Jayasuriya, N., and Rathnayake, N. (2025). Bridging the Chatbot Connection: The Role of AI-Driven Chatbot Affordances in e-Commerce Purchase Intentions. Int. J. Sociol. Soc. Policy, 1–21. sourceNot checked — no full text of the source
• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceCould not verify

  The supplied passages are repository metadata, not findings from the paper’s full text. Searches for reactance, cognitive or affective overload, satisfaction, and downstream evaluations did not retrieve a substantive full-text passage that can verify the claim. (LLM, conf=0.72)

  Experimental — accuracy not yet measured
• cites [44] Vebrianti, 2025, AI Chatbots in E-Commerce: Enhancing Customer Engagement, Satisfaction and Loyalty, PaperASIA sourcePartly supported

  Section 2.5 says that chatbots requiring high cognitive effort can harm user experience and reduce adoption, and that unmet expectations may cause dissatisfaction. Section 2.7 links dissatisfaction with poorer customer loyalty. The copy does not identify reactance or affective overload as the mechanisms behind these effects; page numbers were not provided. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [48] Gao, 2025, The Influence of Artificial Intelligence Chatbot Problem Solving on Customers’ Continued Usage Intention in e-Commerce Platforms: An Expectation-Confirmation Model Approach, J. Bus. Res. sourceNot checked — no full text of the source
• cites [50] Sundjaja, 2025, The Determinant Factors of Continuance Use of Customer Service Chatbot in Indonesia E-Commerce: Extended Expectation Confirmation Theory, J. Sci. Technol. Policy Manag. sourceNot checked — no full text of the source
• cites [61] Park, S., Jung, H., Ryskeldiev, B., Lee, J., Lee, G., Jeong, H., and Chun, M. (2024). Toward Closed-Domain Conversational Item Listing Assistant for Improvement of Experiences of Older Adults in Customer-to-Customer (C2C) Marketplaces. Proceedings of the ASSETS 2024—Proceedings of the 26th International ACM SIGACCESS Conference on Computers and Accessibility, Association for Computing Machinery, Inc. sourceNot checked — no full text of the source
• cites [64] Li, 2025, How Do Artificial Intelligence Chatbots Affect Customer Purchase? Uncovering the Dual Pathways of Anthropomorphism on Service Evaluation, Inf. Syst. Front. sourceNot checked — no full text of the source
• cites [69] Jia, 2025, A Study on the Factors That Influence Consumers’ Continuance Intention to Use Artificial Intelligence Chatbots in a Pharmaceutical e-Commerce Context, Electron Libre sourceNot checked — no full text of the source

May combine two claims.

Observed conversion/purchase behavior is positive when friction drops or recommendations fit, but not universal.

“Positive when fric-tion drops or rec-ommendations fit, but not universal”

• cites [10] Nie, G., Zhi, R., Yan, X., Du, Y., Zhang, X., Chen, J., Zhou, M., Chen, H., Li, T., and Cheng, Z. (2024). A Hybrid Multi-Agent Conversational Recommender System with LLM and Search Engine in E-Commerce. Proceedings of the RecSys 2024—Proceedings of the 18th ACM Conference on Recommender Systems, Association for Computing Machinery, Inc. sourceNot checked — no full text of the source
• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source
• cites [41] Frank, 2025, Consumer Acceptance of High-Autonomy AI Assistants Is Driven by Perceived Benefits in Online Shopping Settings Characterized by Scarcity, Psychol. Mark. sourcePartly supported

  The Results (§6.2) report higher ad click-through for the high-autonomy AI assistant, and the Limitations (§11.3) say the adoption benefit is confined to scarcity settings, supporting a positive but non-universal response. The paper does not establish observed conversions or purchases caused by reduced friction or better-fitting recommendations; §11.3 explicitly calls for future studies tracking actual purchase behavior. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [45] Ranjan, N., Gupta, A., Sharma, L.K., and Kapoor, S. (2024). Adoption of AI Chatbots: Revolutionizing E-Grocery Shopping in Delhi-NCR. Proceedings of the 2024 International Conference on Artificial Intelligence and Emerging Technology, Global AI Summit 2024, Institute of Electrical and Electronics Engineers Inc. sourceNot checked — no full text of the source
• cites [55] Ovalle, C. (2024). Integrated Recommendation System in ChatGPT to Analyze Post-Purchase Behavior of E-Commerce Store Users. Proceedings of the LACCEI International Multi-Conference for Engineering, Education and Technology, Latin American and Caribbean Consortium of Engineering Institutions. sourcePartly supported

  In E. Product Completion (page unavailable), the paper reports higher repurchase rates and purchases of recommended products among users who interacted with its ChatGPT recommendation system. It also describes limitations of recommendations that fail to meet users’ needs, consistent with the claim that positive effects are not universal. The paper does not assess whether reducing purchase friction improves conversion or purchasing behavior. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [64] Li, 2025, How Do Artificial Intelligence Chatbots Affect Customer Purchase? Uncovering the Dual Pathways of Anthropomorphism on Service Evaluation, Inf. Syst. Front. sourceNot checked — no full text of the source
• cites [65] Kumar, 2025, Examining Humanized Bot Interaction for the Retail Revenue Growth, J. Revenue Pricing Manag. sourceNot checked — no full text of the source
• cites [68] Becan, 2025, How Technology Readiness Influences Behavioral and Purchasing Intention: Serial Multiple Mediating Role of Attitude toward AI and AI-Driven Consumer Chatbot Experience, Digit. Transform. Soc. sourceNot checked — no full text of the source

May combine two claims.

Satisfaction/CX/perceived service quality is mostly positive, but design can introduce negative affect (reactance, overload).

“Mostly positive, but design can introduce negative affect (reactance, overload)”

• cites [4] 2024, Purchase Intentions in a Chatbot Environment: An Examination of the Effects of Customer Experience, Oeconomia Copernic. sourcePartly supported

  The Results (page unavailable) report positive effects of entertainment, media appeal and social presence on customer experience, alongside a negative effect of privacy risk. The paper also discusses negative customer experiences from poor chatbot information in the Information section, but the passages do not establish design-induced reactance or overload. (LLM, conf=0.83)

  Experimental — accuracy not yet measured
• cites [7] Klein, 2023, The Impact of Anthropomorphism on Customer Satisfaction in Chatbot Commerce: An Experimental Study in the Food Sector, Electron. Commer. Res. sourcePartly supported

  The Results (Correlation analysis; page unavailable) report a positive relationship between anthropomorphic design cues and customer satisfaction. The Literature review (Chatbot commerce; page unavailable) also warns that highly human-like agents can raise expectations and leave customers disappointed. However, the paper does not identify reactance as an effect, and its Anthropomorphism section says such design may help users cope with information overload rather than introduce it. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites [22] Yatawara, K.M.B., Sampath, T., Kalupahana, P.L., Rathnayake, S.O., Jayasuriya, N., and Rathnayake, N. (2025). Bridging the Chatbot Connection: The Role of AI-Driven Chatbot Affordances in e-Commerce Purchase Intentions. Int. J. Sociol. Soc. Policy, 1–21. sourceNot checked — no full text of the source
• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceCould not verify

  The indexed passages contain a repository listing and a link to the paper’s PDF, but no substantive passage from the paper. Searches for satisfaction, service quality, reactance, and overload found nothing that addresses whether chatbot experiences are mostly positive or whether design introduces negative affect. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [44] Vebrianti, 2025, AI Chatbots in E-Commerce: Enhancing Customer Engagement, Satisfaction and Loyalty, PaperASIA sourcePartly supported

  The Discussion reports positive effects of chatbot features on customer satisfaction. Section 2.5 also says that overly complex interfaces can frustrate users and that high cognitive effort can harm the user experience. These passages support the general contrast, but the paper does not establish reactance or overload as specific negative responses, nor does it show that customer responses are mostly positive overall. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites [48] Gao, 2025, The Influence of Artificial Intelligence Chatbot Problem Solving on Customers’ Continued Usage Intention in e-Commerce Platforms: An Expectation-Confirmation Model Approach, J. Bus. Res. sourceNot checked — no full text of the source
• cites [50] Sundjaja, 2025, The Determinant Factors of Continuance Use of Customer Service Chatbot in Indonesia E-Commerce: Extended Expectation Confirmation Theory, J. Sci. Technol. Policy Manag. sourceNot checked — no full text of the source
• cites [61] Park, S., Jung, H., Ryskeldiev, B., Lee, J., Lee, G., Jeong, H., and Chun, M. (2024). Toward Closed-Domain Conversational Item Listing Assistant for Improvement of Experiences of Older Adults in Customer-to-Customer (C2C) Marketplaces. Proceedings of the ASSETS 2024—Proceedings of the 26th International ACM SIGACCESS Conference on Computers and Accessibility, Association for Computing Machinery, Inc. sourceNot checked — no full text of the source
• cites [64] Li, 2025, How Do Artificial Intelligence Chatbots Affect Customer Purchase? Uncovering the Dual Pathways of Anthropomorphism on Service Evaluation, Inf. Syst. Front. sourceNot checked — no full text of the source
• cites [69] Jia, 2025, A Study on the Factors That Influence Consumers’ Continuance Intention to Use Artificial Intelligence Chatbots in a Pharmaceutical e-Commerce Context, Electron Libre sourceNot checked — no full text of the source

May combine two claims.

Satisfaction rises via problem-solving, confirmation, usefulness, and interaction/service quality.

“Satisfaction rises via problem-solving, confirmation, usefulness, and interaction/service quality.”

• cites [4] 2024, Purchase Intentions in a Chatbot Environment: An Examination of the Effects of Customer Experience, Oeconomia Copernic. sourcePartly supported

  The Information section says information quality is positively related to chatbot satisfaction, and the Discussion reports a positive relationship between media appeal and satisfaction. A study summarized in the Limitations and directions for future research section includes problem solving and interaction among marketing efforts linked indirectly to satisfaction. The paper does not establish confirmation or usefulness as drivers of satisfaction; its discussion of service quality concerns customer experience instead. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites [7] Klein, 2023, The Impact of Anthropomorphism on Customer Satisfaction in Chatbot Commerce: An Experimental Study in the Food Sector, Electron. Commer. Res. sourcePartly supported

  The Literature review (Chatbot commerce; page unavailable) says that accurate recommendations and an efficient, convenient chatbot experience contribute to customer satisfaction. The Method (Measurement; page unavailable) also includes whether expectations are met in its satisfaction measure. But the paper does not establish problem-solving, confirmation, usefulness, and interaction/service quality as the pathways by which satisfaction rises; its Discussion instead identifies perceived enjoyment, attitude, and trust as mediators. (LLM, conf=0.83)

  Experimental — accuracy not yet measured
• cites [22] Yatawara, K.M.B., Sampath, T., Kalupahana, P.L., Rathnayake, S.O., Jayasuriya, N., and Rathnayake, N. (2025). Bridging the Chatbot Connection: The Role of AI-Driven Chatbot Affordances in e-Commerce Purchase Intentions. Int. J. Sociol. Soc. Policy, 1–21. sourceNot checked — no full text of the source
• cites [24] Pizzi, 2021, Artificial Intelligence and the New Forms of Interaction: Who Has the Control When Interacting with a Chatbot?, J. Bus. Res. sourceCould not verify

  The supplied passages and searches for satisfaction, problem-solving, confirmation, usefulness, and service quality return only an IRIS repository record. That record lists an accepted-manuscript PDF but contains no passage from the paper addressing what raises satisfaction, so the claim cannot be checked against our copy. (LLM, conf=0.74)

  Experimental — accuracy not yet measured
• cites [44] Vebrianti, 2025, AI Chatbots in E-Commerce: Enhancing Customer Engagement, Satisfaction and Loyalty, PaperASIA sourcePartly supported

  The Discussion (§4.3; page unavailable) links satisfaction to useful solutions, information quality, and better chatbot service; §2.6 also describes the importance of relevant responses and customer expectations. The paper does not identify confirmation as a distinct driver of satisfaction, and it does not test interaction or service quality as separate variables. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites [48] Gao, 2025, The Influence of Artificial Intelligence Chatbot Problem Solving on Customers’ Continued Usage Intention in e-Commerce Platforms: An Expectation-Confirmation Model Approach, J. Bus. Res. sourceNot checked — no full text of the source
• cites [50] Sundjaja, 2025, The Determinant Factors of Continuance Use of Customer Service Chatbot in Indonesia E-Commerce: Extended Expectation Confirmation Theory, J. Sci. Technol. Policy Manag. sourceNot checked — no full text of the source
• cites [61] Park, S., Jung, H., Ryskeldiev, B., Lee, J., Lee, G., Jeong, H., and Chun, M. (2024). Toward Closed-Domain Conversational Item Listing Assistant for Improvement of Experiences of Older Adults in Customer-to-Customer (C2C) Marketplaces. Proceedings of the ASSETS 2024—Proceedings of the 26th International ACM SIGACCESS Conference on Computers and Accessibility, Association for Computing Machinery, Inc. sourceNot checked — no full text of the source
• cites [64] Li, 2025, How Do Artificial Intelligence Chatbots Affect Customer Purchase? Uncovering the Dual Pathways of Anthropomorphism on Service Evaluation, Inf. Syst. Front. sourceNot checked — no full text of the source
• cites [69] Jia, 2025, A Study on the Factors That Influence Consumers’ Continuance Intention to Use Artificial Intelligence Chatbots in a Pharmaceutical e-Commerce Context, Electron Libre sourceNot checked — no full text of the source

May combine two claims.

High-autonomy cues can reduce intention via powerlessness/threat-to-freedom, unless urgency/scarcity raises benefit salience and attenuates resistance.

“High-autonomy cues can reduce intention via powerless-raises benefit salience and attenuates resistance. ness/threat-to-freedom, unless urgency/scarcity”

• cites [4] 2024, Purchase Intentions in a Chatbot Environment: An Examination of the Effects of Customer Experience, Oeconomia Copernic. sourceCould not verify

  The Results (page unavailable) report that privacy risk negatively affects purchase intention, while the Media Appeal section discusses technology anxiety and chatbot experience. Neither passage addresses high-autonomy cues, powerlessness or threat to freedom, or whether urgency or scarcity attenuates resistance; searches for those terms found no supporting passage. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [6] Qi, 2025, AI Chatbots with Visual Search: Impact on Luxury Fashion Shopping Behavior, J. Glob. Sch. Mark. Sci. Bridg. Asia World sourceNot checked — no full text of the source
• cites [8] Schindler, 2024, How Speaking versus Writing to Conversational Agents Shapes Consumers’ Choice and Choice Satisfaction, J. Acad. Mark. Sci. sourceCould not verify

  The Mediation and Manipulation check passages examine speaking versus writing to conversational agents, brand equity, and intention to choose. Searches for autonomy, powerlessness or threat to freedom, and urgency or scarcity found no passage addressing the claimed mechanism or exception. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [22] Yatawara, K.M.B., Sampath, T., Kalupahana, P.L., Rathnayake, S.O., Jayasuriya, N., and Rathnayake, N. (2025). Bridging the Chatbot Connection: The Role of AI-Driven Chatbot Affordances in e-Commerce Purchase Intentions. Int. J. Sociol. Soc. Policy, 1–21. sourceNot checked — no full text of the source
• cites [30] Hassan, 2025, The Moderating Role of Personalized Recommendations in the Trust–Satisfaction–Loyalty Relationship: An Empirical Study of AI-Driven e-Commerce, Future Bus. J. sourceCould not verify

  The Summary of the literature review (page unavailable) discusses control of chatbot interactions, and Structural equation modelling (SEM; page unavailable) reports a possible weakening of the trust–satisfaction relationship through personalization. Neither addresses high-autonomy cues reducing intention through powerlessness or threats to freedom, or urgency and scarcity attenuating resistance. Searches for those terms and their paraphrases found no supporting passage. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [31] Ding, Y., and Najaf, M. (2024). Interactivity, Humanness, and Trust: A Psychological Approach to AI Chatbot Adoption in e-Commerce. BMC Psychol., 12. sourceCould not verify

  The Discussion (page not provided) examines trust as a mediator and perceived enjoyment as a moderator of chatbot adoption. Searches for autonomy, powerlessness, threat to freedom, urgency, and scarcity found no passage addressing the claimed mechanism or exception. (LLM, conf=0.72)

  Experimental — accuracy not yet measured
• cites [32] Song, 2024, Uncanny Valley Effects on Chatbot Trust, Purchase Intention, and Adoption Intention in the Context of E-Commerce: The Moderating Role of Avatar Familiarity, Int. J. Hum. Comput. Interact. sourceCould not verify

  The Discussion (section 6.1; page unavailable) links avatar eeriness to lower chatbot trust, and section 3.4 hypothesizes effects on purchase and reuse intentions. Neither passage addresses high-autonomy cues, powerlessness or threats to freedom, or urgency and scarcity attenuating resistance; searches for those terms found no supporting passage in our copy. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [39] Liu, J., and Chen, J. (2025). Chatbot-Aided Product Purchases among Generation Z: The Role of Personality Traits. Front. Psychol., 16. sourceCould not verify

  The Results (Results of NCA; page unavailable) discuss conditions for purchase intention, while the conceptual background discusses chatbot information cues. Neither passage addresses high-autonomy cues, powerlessness or threats to freedom, or whether urgency or scarcity attenuates resistance; searches for those terms found no relevant passage. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source
• cites [41] Frank, 2025, Consumer Acceptance of High-Autonomy AI Assistants Is Driven by Perceived Benefits in Online Shopping Settings Characterized by Scarcity, Psychol. Mark. sourceSupported

May combine two claims. The quote only partly states this claim.

Brand choice/choice share effects are context-dependent; effects are stronger when modality matches product type and when brand equity is low.

“Context-dependent; comes”

• cites [8] Schindler, 2024, How Speaking versus Writing to Conversational Agents Shapes Consumers’ Choice and Choice Satisfaction, J. Acad. Mark. Sci. sourcePartly supported

  The Study 5 findings report a stronger positive effect of matching communication modality to product type on intention to choose a low-equity brand, specifically when the alternatives differ in brand equity. The Method section shows that participants rated their intention to choose; it does not establish an effect on observed brand-choice share. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

May combine two claims.

Behavioral lift appears when systems reduce friction (latency/steps) or improve option fit (recommendation relevance).

“Behavioral lift appears when systems reduce fric-havioral intention rather than verified execution. tion (latency/steps) or improve option fit (recom-mendation relevance).”

• cites [10] Nie, G., Zhi, R., Yan, X., Du, Y., Zhang, X., Chen, J., Zhou, M., Chen, H., Li, T., and Cheng, Z. (2024). A Hybrid Multi-Agent Conversational Recommender System with LLM and Search Engine in E-Commerce. Proceedings of the RecSys 2024—Proceedings of the 18th ACM Conference on Recommender Systems, Association for Computing Machinery, Inc. sourceNot checked — no full text of the source
• cites [40] Frank, 2024, Autonomy, Power and the Special Case of Scarcity: Consumer Adoption of Highly Autonomous Artificial Intelligence, Br. J. Manag. sourceNot checked — no full text of the source
• cites [41] Frank, 2025, Consumer Acceptance of High-Autonomy AI Assistants Is Driven by Perceived Benefits in Online Shopping Settings Characterized by Scarcity, Psychol. Mark. sourcePartly supported

  Section 4 suggests that fast, efficient decisions can make high-autonomy AI appealing in scarcity-related shopping, and Section 2 reports that preference-matched recommendations are associated with greater satisfaction than mismatched ones. But the paper does not show behavioral lift from reduced latency or fewer steps, or from improved recommendation relevance. Its own results (Section 10.2.2) identify experiential benefits—not functional benefits—as the significant pathway to adoption under scarcity. (LLM, conf=0.83)

  Experimental — accuracy not yet measured
• cites [45] Ranjan, N., Gupta, A., Sharma, L.K., and Kapoor, S. (2024). Adoption of AI Chatbots: Revolutionizing E-Grocery Shopping in Delhi-NCR. Proceedings of the 2024 International Conference on Artificial Intelligence and Emerging Technology, Global AI Summit 2024, Institute of Electrical and Electronics Engineers Inc. sourceNot checked — no full text of the source
• cites [55] Ovalle, C. (2024). Integrated Recommendation System in ChatGPT to Analyze Post-Purchase Behavior of E-Commerce Store Users. Proceedings of the LACCEI International Multi-Conference for Engineering, Education and Technology, Latin American and Caribbean Consortium of Engineering Institutions. sourcePartly supported

  The paper’s E. Product Completion section reports higher repurchase rates and more purchases of recommended products among users of the ChatGPT-integrated recommendation system. It also links personalized, relevant recommendations to repeat purchases. It does not show that reducing latency or the number of steps produces a behavioral lift. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [64] Li, 2025, How Do Artificial Intelligence Chatbots Affect Customer Purchase? Uncovering the Dual Pathways of Anthropomorphism on Service Evaluation, Inf. Syst. Front. sourceNot checked — no full text of the source
• cites [65] Kumar, 2025, Examining Humanized Bot Interaction for the Retail Revenue Growth, J. Revenue Pricing Manag. sourceNot checked — no full text of the source
• cites [68] Becan, 2025, How Technology Readiness Influences Behavioral and Purchasing Intention: Serial Multiple Mediating Role of Attitude toward AI and AI-Driven Consumer Chatbot Experience, Digit. Transform. Soc. sourceNot checked — no full text of the source

May combine two claims.

Many studies rely on simulated systems or intention data, whereas observed delegation behavior, transaction execution, and longitudinal dynamics remain comparatively scarce.

“Many studies rely on simulated systems or intention data, whereas observed delegation behavior, transaction execution, and longitudinal dynamics remain comparatively scarce.”

Probably not a substantial claim: may not pass S4 (based).

Urgency can attenuate governance resistance.

“urgency can attenuate governance resistance”

May combine two claims.

Most outcomes are intention- or vignette-based, so calibration, verification, and error-recovery behaviors remain under-measured.

“Because most outcomes are intention-or vignette-based, calibration, verification, and error-recovery behaviors remain under-measured.”

May combine two claims.

Many studies address conversational commerce and decision support (often measured via intentions), whereas fewer studies test observable delegation behaviors or execution outcomes.

“many studies address conversational commerce and decision support (often measured via intentions), whereas fewer studies test observable delegation behaviors or execution outcomes”

May combine two claims.

Consumer-level outcomes are more mixed, showing that there are conflicts between convenience and freedom.

“Consumer-level outcomes are more mixed, showing that there are conflicts between convenience and freedom.”

Probably not a substantial claim: may not pass S4 (based).

Constructs central to safe and robust delegation, trust calibration, verification behavior, recovery after error, and post-delegation regret/blame, are frequently theorized but inconsistently operationalized.

“Constructs central to safe and robust delegation, trust calibration, verification behavior, recovery after error, and post-delegation regret/blame, are frequently theorized but inconsistently operationalized.”

Probably not a substantial claim: may not pass S4 (based). May combine two claims.

Most of the research emphasizes self-reported intentions, examines assistive rather than truly agentic systems, and ignores the socio-technical relationships between autonomy, task scope, and transaction capability.

“Most of the research emphasizes self-reported intentions, examines assistive rather than truly agentic systems, and ignores the socio-technical relationships between autonomy, task scope, and transaction capability.”

May combine two claims.

While functional acceptance is confirmed by these studies, their empirical focus largely reflects assistive autonomy (information and persuasion) and true autonomy and complete task completion are rarely considered or empirically validated.

“While functional acceptance is confirmed by these studies, their empirical focus largely reflects assistive autonomy (information and persuasion) and true autonomy and complete task completion are rarely considered or empirically validated.”

May combine two claims.

Much of the evidence operationalizes delegation indirectly via satisfaction or intention constructs rather than via observed reliance behaviors, which limits inference about calibrated delegation under real execution risk.

“much of this evidence operationalizes delegation indirectly via satisfaction or intention constructs rather than via observed reliance behaviors, which limits inference about calibrated delegation under real execution risk”

May combine two claims.

Most of the research is still self-reported and cross-sectional, offering little insight into how delegation changes with repeated interaction or actual transactional exposure.

“most of the research is still self-reported and cross-sectional, offering little insight into how delegation changes with repeated interaction or actual transactional exposure”

May combine two claims.

The mapped literature is concentrated in text-based, low-autonomy assistants focused on recommendation and post-purchase support; coverage drops sharply for workflow-level autonomy, cart building, checkout/payment execution, and negotiation.

“The mapped literature is concentrated in text-based, low-autonomy assistants focused on recommendation and post-purchase support; coverage drops sharply for workflow-level autonomy, cart building, checkout/payment execution, and negotiation.”

May combine two claims.

Consumer outcomes are not as consistently covered as marketing outcomes, and their measurement is often indirect.

“Consumer outcomes are not as consistently covered as marketing outcomes, and measurement is often indirect.”

May combine two claims.

Task scope clusters around pre-purchase decision support and post-purchase service, with limited coverage of true execution.

“Task scope clusters around pre-purchase decision support and post-purchase service, with limited coverage of true execution.”

May combine two claims.

A second cluster centers on perceived control (n=4), perceived autonomy/agency (n=5), and reactance/powerlessness (n=4).

“A second cluster centers on perceived control (n = 4), perceived autonomy/agency (n = 5), and reactance/powerlessness (n = 4).”

May combine two claims.

The majority of interaction is text-based chat, usually user-initiated.

“The majority of interaction is text-based chat, usually user-initiated (e.g., [42,46,48,55]).”

May combine two claims.

Evidence with more behavior-proximal outcomes is available but relatively scarce in the mapped set.

“Evidence with more behavior-proximal outcomes is available but is relatively scarce in the mapped set (e.g., [10,38,53,55]).”

May combine two claims.

Reliance generally escalates with trust and diminishes with resistance.

“reliance generally escalates with trust and diminishes with resistance”

May combine two claims.

The mapped corpus (N=47 studies) is geographically concentrated in East and Southeast Asia (combined 42.5%), with China the single most common country setting (19.1%).

“As shown in Table 4, the mapped corpus (N = 47 studies) is geographically concentrated in East and Southeast Asia (combined 42.5%), with China representing the single most common country setting (19.1%).”

May combine two claims.

Privacy concern (n=4) and perceived risk (n=3) manifest less frequently than usefulness/trust and exhibit more heterogeneous effects.

“Privacy concern (n = 4) and perceived risk (n = 3) manifest less frequently than usefulness/trust and exhibit more heterogeneous effects.”

May combine two claims.

Cues of urgency or stakes can make people less resistant to governance, weakening the autonomy → powerlessness pathway and making higher autonomy more acceptable in situations where people need it.

“Second, cues of urgency or stakes, such as scarcity and time pressure, can make people less resistant to governance.This weakens the autonomy → powerlessness pathway and makes higher autonomy more acceptable in situations where people need it.”

May combine two claims.

Moderation and boundary conditions are reported inconsistently, yet several recurring themes surface.

“Moderation and boundary conditions are reported inconsistently (moderators tested: varied across determinants), yet several recurring themes surface.”

May combine two claims.

Anthropomorphism is common but conceptually distinct from autonomy; numerous studies use it as a manipulation or measured cue.

“Anthropomorphism is common but conceptually distinct from autonomy; numerous studies use it as a manipulation or measured cue, frequently through avatars, naming, warmth cues, or social presence [24,38,[63][64][65].”

May combine two claims.

Level 3 (high autonomy with execution) is infrequent and predominantly realized through experimental frameworks that alter control rights and execution.

“Level 3 (high autonomy with execution) is infrequent and predominantly realized through experimental frameworks that distinctly alter control rights and execution [40,41,46].”

May combine two claims.

The evidence base is dominated by intention-only/cross-sectional designs (55.3%), while behavioral evidence in realistic contexts remains comparatively scarce (14.9%).

“In parallel, the evidence base is dominated by intention-only/cross-sectional designs (55.3%), while behavioral evidence in realistic contexts remains comparatively scarce (14.9%).”

May combine two claims.

The high-autonomy manipulation studies focus on full execution (order + payment), which makes them crucial but also shows how infrequently execution is seen in naturalistic commerce use.

“The high-autonomy manipulation studies [40,41] focus on full execution (order + payment), which makes them crucial but also shows how infrequently execution is seen in naturalistic commerce use.”

May combine two claims.

Sample composition includes a substantial share of online panels (23.4%) and student samples (21.3%), alongside customer/user samples (36.2%).

“Sample composition further reflects typical generalizability constraints in this literature, with a substantial share of online panels (23.4%) and student samples (21.3%), alongside customer/user samples (36.2%).”

May combine two claims.

The positive marketing benefits of agentic or assistant-enabled commerce include higher purchase/adoption intention, higher observed conversion/acceptance, higher spend/GMV, and stronger retention/continuance/loyalty.

“These benefits are seen as higher purchase/adoption intention, higher observed conversion/acceptance, higher spend/GMV, and stronger retention/continuance/loyalty.”