Benchmarking consumer behaviour using the generative AI: a systematic review and theoretical critique

Purpose The study is a systematic and critical analysis of the influence of generative artificial intelligence on consumer behaviour. It investigates the specific effects of purchase intention on the influence of psychological drivers, trust, cultural influences and digital literacy through theory of planned behaviour (TPB), unified theory of acceptance and use of technology (UTAUT) and diffusion of innovations (DOI). Although the review has been taken in a global view and with contextual insight into emerging markets where evidence supports the same, the same is discussed. Design/methodology/approach The article is based on the PRISMA 2020 methodology and summarises 25 peer-reviewed journal articles published between 2018 and 2026. Structured Boolean queries were used in the search in Scopus, Web of Science, Google Scholar and Emerald. Methodological rigour was achieved through a dual-reviewer screening process and a CASP quality appraisal. Findings were synthesised using thematic analysis. Findings The findings show that large language models, image generators, and conventional AI systems (e.g. recommender systems and predictive analytics) have a substantial impact on consumer cognition, emotions, and decision-making. Among 25 studies reviewed, a reduced percentage are those that specifically analyse generative AI systems, and most of the reviews analyse traditional AI applications, an emerging yet developing evidence base. Generative AI indirectly influences purchase intention, mediated by attitude, perceived behavioural control and social influence, with trust as a key mediator. Adoption is further influenced by cultural context and digital literacy, as it not only serves as a boundary condition that impacts trust formation and behavioural results. Research limitations/implications The research also relies on secondary data, and this could restrict its generalisability in different contexts. Moreover, generative AI technologies are changing rapidly, which might affect the temporal relevance of findings. Practical implications In the banking industry, Gen AI chatbots can help in investment guidance, personalised financial education and fraud alerts. In the education sector, AI tutors and content generators offer vernacular, personalised learning support. The overall acceptance of this technology depends upon the perceived capability, accuracy, trust and parents' influence. From the retail perspective, AI-enabled customisation, virtual try-ons and in-store smart kiosks are very effective. Originality/value This research is novel as it incorporates TPB, UTAUT and DOI into a single model that separates between the generative and traditional AI as a means of influencing consumer behaviour. It presents a framework for governance that incorporates responsible research and innovation (RRI), which connects ethical implications to mechanisms of trust and adoption in AI-driven consumer contexts.