Background

Agentic AI is currently among the most loudly promoted technology categories in retail. Software vendors ship agents for merchandising, fulfillment, customer service, and product onboarding; consultancies publish adoption figures; trade press reports a transformation underway. The peer-reviewed evidence base, by contrast, is remarkably thin: academic work on AI in retail largely predates the agent wave, and the few studies at the intersection are preprints reporting single-site proofs of concept.

Practitioner surveys themselves suggest a gap. Industry studies report that a large majority of organizations say they use AI agents, while only a small fraction of agentic use cases reach production. Analysts have coined the term “agent washing” for products marketed as agentic that automate conventional workflows.

Research in information systems has shown that discussions about technology often follow fashion waves, with practitioners and academics influencing each other. It has been suggested that IS scholars should take a more active role in shaping this discourse rather than just following trends. The example of agentic AI in retail provides an ideal opportunity to do so.

Object of investigation

This thesis takes the divergence between promise, evidence, and practice as its core idea. It systematically compares three discourses: peer-reviewed academic literature, practitioner and vendor grey literature, and reported practice in retail organizations. On this basis, the thesis derives research directions that follow demonstrably from the identified gaps. These should be presented as substantiated propositions, each traceable to specific evidence in the analyzed corpus.

Following a multivocal literature review, claims are to be identified and analyzed. Depending on scope, the review can be complemented by interviews with practitioners in retail organizations in order to contrast the published discourse with what is actually in operation.

Possible research questions

• Which agentic AI applications in retail are promoted in practitioner and vendor sources, and which of them are addressed by academic research?
• What kind and quality of evidence underpins the claims made in each of these discourses?
• Where do promises, research findings and reported practice diverge, and how can these divergences be explained?

Possible results

• A multivocal literature review with a structured claim-and-evidence matrix by retail process area.
• An assessment of the maturity of agentic AI in retail that distinguishes what is evidenced from what is asserted.
• A research agenda and/or practical implications for retailers assessing vendor offerings.

Starting literature

• Baskerville, R. L. & Myers, M. D. (2009). Fashion Waves in Information Systems Research and Practice. MIS Quarterly, 33(4), 647–662. DOI 10.2307/20650319.
• Vu, H., Klievtsova, N., Leopold, H., Rinderle-Ma, S., Kampik, T. (2026). Agentic Business Process Management: Practitioner Perspectives on Agent Governance in Business Processes. DOI 10.1007/978-3-032-02936-2_3
• Dumas, M., Fournier, F., Limonad, L. et al. (2023). AI-augmented Business Process Management Systems: A Research Manifesto. ACM Transactions on Management Information Systems, 14(1), Art. 11. DOI 10.1145/3576047.
• Cao, L. (2021). Artificial intelligence in retail: applications and value creation logics. International Journal of Retail & Distribution Management, 49(7), 958–976. DOI 10.1108/IJRDM-09-2020-0350
• Weber, F. D., Schütte, R. (2019). State-of-the-art and adoption of artificial intelligence in retailing. Digital Policy, Regulation and Governance, 21(3), 264–279. DOI 10.1108/DPRG-09-2018-0050.

Recommended prior knowledge

An interest in retail and in AI agents. Empirical research methods are useful if the interview component is chosen.