
“A year spent in artificial intelligence is enough to make one believe in God.”
– Alan Jay Perlis, first ever Turing Award recipient
Marcel Pisu & Stefan Kluge
Artificial Intelligence (AI) is probably the most frequently used buzzword in every business meeting. Unfortunately, AI often remains a buzzword because companies are not always able to use it to profitably. However, there exist enumerable use cases for AI, especially in the field of marketing. In response to changing customer behavior, marketers need to strategically employ AI along the customer journey to develop longer-lasting and more value-generating customer relationships (Huang and Rust 2023). Herefore, generative AI, such as chatbots, plays a crucial role in building a personal relationship with each customer (Li, Yao, and Nan 2023).
Generally, it can be said that AI offers benefits for both the customer by enhancing the customer experience and the company through optimizing operational processes and increasing efficiency. Many companies, such as Netflix, Amazon, or Under Armour, already make use of this technology. However, customers can have several considerations towards the use of AI which potentially influences their technology acceptance. Individuals differ in their readiness for technology, which affects whether customers accept and even endorse the use of technology in the customer journey or neglect it. Consequently, regardless the configuration or technical capabilities of a chatbot, for example, this individual preposition can influence a customer’s engagement behavior (Yin, Li, and Qiu 2023). However, there are two factors that independently influence people’s technology acceptance, namely perceived usefulness and perceived ease of use. Perceived usefulness refers to the extent to which the individual believes that using the technology would be beneficial. Perceived ease of use refers to the effort that is considered necessary to use the technology (Davis 1985). More recent studies indicate that there are more factors that influence customers’ perception and acceptance of AI:
Safety
First, individuals are seen to consider ethical factors and the safety of using the technology (Arango, Singaraju, and Niininen 2023; Hui et al. 2023; Li, Yao, and Nan 2023). They are sensitive to the type of information that is being processed by AI, i.e. individuals are less sensitive to the disclosure of information such as name or date of birth than to the disclosure of their ID number (Li, Yao, and Nan 2023). Moreover, if AI is used for ethical reasons, e.g., by generating artificial instead of real images of children to protect their privacy rights, individuals are more likely to support the usage of technology (Arango, Singaraju, and Niininen 2023).
Environment
Second, the setting in which AI is used and the emotional state of individuals matter (Budler, Gosak, and Stiglic 2023; Crolic et al. 2022; Prentice and Nguyen 2020). Generally, conversational AI is perceived user-friendly and useful (Budler, Gosak, and Stiglic 2023). This also applies to retail, as AI-supported shopping in stores is positively correlated with the service experience (Farooqui 2022). Yet in “people-intensive industries such as hotels” (Prentice and Nguyen 2020) customers tend to prefer to be served by people rather than AI (Prentice and Nguyen 2020). Moreover, for customers that are in an angry emotional state, anthropomorphic AI chatbots, for example, can reduce customer satisfaction and therefore the likelihood of purchase (Crolic et al. 2022). As a countermeasure, Crolic et al. propose that AI should be used to recognize whether a customer is in an angry or non-angry emotional state, although there is a considerable probability that the customer’s emotional state will be misidentified, especially if the customer is dishonest or expresses vaguely (Huang and Rust 2023).
Enjoyment
Last, the configuration and empathy of conversational AI influences the way individuals perceive it (Elmashhara et al. 2024; Hui et al. 2023; Kim, Kim, and Baek 2024; Li, Yao, and Nan 2023; Puntoni et al. 2021). To increase customer engagement, chatbots that are able to create an emotional connection tend to increase positive customer engagement behavior (Li, Yao, and Nan 2023). Further, Kim, Kim, and Baek (2024) propose that “AI chatbots should be crafted with a focus on enjoyable, user-centered interfaces that foster long-term user satisfaction and engagement”. In this context, Elmashhara et al. (2024) found that gamification of chatbots with simple games of chance, for example flipping a coin to win a discount, can increase both customer engagement and purchase outcomes
To conlcude, AI can be both beneficial and harmful in customer service. Customers naturally have different levels of technology readiness which influences the acceptance for the use of AI along the customer journey (Yin, Li, and Qiu 2023). Nevertheless, conversational AI, for example, needs to be configurated individually for different use cases or target groups and requires a certain degree of adaptability (Garvey, Kim, and Duhachek 2023). Additionally, design decisions for AI applications in marketing must be made carefully, with the customer at the center and ethical guidelines in mind (Puntoni et al. 2021).
State-of-the-art AI is not capable of replacing humans entirely. Although the reviewed studies revealed remarkable results of different AI models, most authors indicate the limitations and drawbacks such as inaccuracy or lack of generating new knowledge (Hu et al. 2023; Whiles and Terry 2024). Yet it can be used as an adjunct tool to assist with recurring tasks or to enhance humans with its generative and analytical strengths. According to this finding, marketers should not follow the approach to replace humans with AI but rather to empower and train them to enhance their efficiency which potentially improves customer service experience and quality.