Abstract
This chapter examines the relationship between society and artificial intelligence (AI), emphasizing the factors driving consumer adoption of AI conversational bots. The authors examine how societal norms, past experiences, and trust in technology influence the acceptance and usage of generative-pre trained transformer (GPT) bots. They provide a theoretical framework, integrating key concepts from social influence and technology acceptance theory, to understand the complex dynamics of GPT bot adoption. Conducting a survey, they analyze data from 412 participants in North America to test various hypotheses. The findings broadly support the proposed model, highlighting the significant roles of social norms, word of mouth, and trust in shaping consumer behaviour towards AI conversational bots. However, an intriguing exception is found in the lack of a direct relationship between behavioural intention and actual technology usage, pointing to the need for further investigation into the factors that bridge the gap between the intention to use and the actual use of AI technologies in everyday contexts.
TopIntroduction
The relationship between society and human intellect is intricate and profound, as societal structures and norms are fundamentally shaped by human cognition and decision-making processes. In this context, integrating artificial intelligence (AI) into human cognitive capabilities is a subject of considerable debate and exploration (Russel & Norvig, 2010). While AI presents notable opportunities for augmenting human capabilities, its potential impact on societal progress is complex and multifaceted (Brynjolfsson & McAfee, 2014). The accelerated advancement in Artificial Intelligence (AI) and Natural Language Processing (NLP) technologies have led to the emergence of unique cutting-edge conversational agents such as the Generative Pre-trained Transformer (GPT) (Abdul-Kader, 2021; Kusal et al., 2022; Paul et al., 2023). The GPT is a significant advancement in the natural language processing field, as this technology enables machines to provide humans with a perception of understanding and communicate in a human-like manner (Yenduri, 2023). It gained popularity due to its extensive performance capabilities and practical conversational abilities (Cano, 2023). A recent version of GPT that captured the public's interest is GPT-3. GPT-3 has demonstrated logical and intellectual responses that mimic that of humans, making it a powerful tool for various sectors, formulating ideas for design concepts (Zhu, 2021) and potential implications for healthcare systems (Sezgin, 2022). These advancements represent a profound and transformative trend in integrating complex AI tools into daily consumer experiences, reshaping how consumers interact with technology (Cano, 2023).
A conversational agent is a multimodal system designed to imitate human conversations by employing spoken or written language, typically through internet-based interactions (Adamopoulou & Moussiades, 2020). As such, this technology can accomplish tasks comparable to the work of many human employees, resulting in possible significant cost savings for organizations (Kusal et al., 2022). The significant capabilities of conversational agents have allowed many organizations and businesses to adopt them into practice. Healthcare, customer services, e-commerce, education, media, and manufacturing are just a few examples (Kustal et al., 2022). The chatbot market saw over a 20% compound annual growth rate increase in 2023 and is expected to grow to a market size of 32.4 billion USD by 2032 (DataHorizzon Research, 2023).
Key Terms in this Chapter
Trust in Technology: The degree of confidence and reliability placed in technology by users. It influences the adoption and usage of technological products and is shaped by WOM and prior experiences (King et al., 2014; Liu & Lopez, 2016).
Conversational Agent: A multimodal system designed to imitate human conversations by employing spoken or written language, typically through internet-based interactions. These agents can perform tasks comparable to many human employees (Adamopoulou & Moussiades, 2020).
Word of Mouth (WOM): The exchange of interpersonal information among adopters and potential adopters of a product or service. In the digital context, WOM leverages online networks to amplify its reach and impact (Cheng & Zhou, 2010; Maxham, 2001).
ChatBot: A specific type of conversational agent that interacts with users primarily through text. It includes retail-service chatbots like SamBot, used for various industry applications (Pradana et al., 2017; Tran et al., 2021).
Social Norms: Established rules that shape behaviour through the encouragement of conformity. They inform group members how to feel and behave in certain situations, exerting social influence on members (Bicchieri & Mercier, 2014; Raven & Rubin, 1976).
ChatGPT: A machine-learning system developed by OpenAI, using generative AI tools to provide conversational responses. It is part of the broader family of GPT models, known for their natural language processing capabilities (van Dis et al., 2023).
Artificial Intelligence (AI): The simulation of human intelligence in machines, characterized by the capability of performing tasks that typically require human cognition (Russel & Norvig, 2010; Brynjolfsson & McAfee, 2014).
Behavioural Intention: An individual's perceived likelihood or propensity to engage in a particular behaviour. It is influenced by the individual's attitude towards the behaviour and subjective norms, including the perceived expectations of significant others. (Ajzen, 1991).
Technology Acceptance Model (TAM): A theoretical model that predicts and explains how users accept and use a technology. The core premise of TAM is that the perceived usefulness and perceived ease of use of technology are primary factors influencing a user's decision to adopt and use that technology. (Davis, 1986).
Natural Language Processing (NLP): A domain of AI that deals with the interaction between computers and human language. It involves the development of algorithms that enable machines to interact, interpret, and generate human-like language (Abdul-Kader, 2021).
Generative Pre-Trained Transformer (GPT): A significant advancement in the natural language processing field, which enables machines to provide humans with a perception of understanding and communicate in a human-like manner. GPT models are pre-trained on extensive text data and fine-tuned for specific tasks (Yenduri, 2023).