This is an excerpt from one of my contributions on the use of responsive chatbots by service businesses. The content was adapted for this blogpost.
(Source: Google Gemini)
Suggested citation: Camilleri, M.A. & Troise, C. (2022). Live support by chatbots with artificial intelligence: A future research agenda. Service Business, https://doi.org/10.1007/s11628-022-00513-9
Chatbots are usually considered as automated conversational systems that are capable of mimicking humanlike conversations. Previous research suggested that, at times, human beings are treating computers as social beings (Nass and Moon 2000; Nass et al. 1994; Rha and Lee 2022) although they are well aware that dialogue programs do not possess emotions, feelings and identities. Individuals may still perceive that service chatbots have some sort of social presence when they interact with them (Leung and Wen 2020; McLean et al. 2020; Pantano & Scarpi, 2022; Schuetzler et al. 2020), even though these technologies are capable of responding to thousands of potential users at once (Caldarini et al. 2022).
Currently, few academic contributions are using theoretical bases like the social presence theory (Grimes et al. 2020; Schuetzler et al. 2020) and/or the social response theory (Adam et al. 2021; Huang and Lin 2011), to explore human-computer interactions, and/or the utility of dialogue systems like chatbots, albeit a few exceptions. A few commentators made specific reference to related theories to describe the characteristics of chatbots or of conversational 4 agents, that are primarily used for consumer engagement purposes (Cheng and Jiang 2020; Kull et al. 2021; Mostafa and Kasamani 2021; Nuruzzaman and Hussain 2020).
The human machine communication theory was formulated in response to the growing number of technologies like AI and robotics, that are designed to function as message sources, rather than as message channels (Flavián et al. 2021). Lewis et al. (2019) contended that social bots, and even a few chatbots have pushed into the realm of what was thought to be a purely human role. Wilkinson et al.’s (2021) study shed light on the human beings’ perceptions about conversational recommender systems. In this case, the authors went on to suggest that experienced users trusted their disruptive technologies and had higher expectations from them.
Other researchers examined the online users’ trust toward chatbots in various settings (Balakrishnan and Dwivedi 2021; Borau et al. 2021; Cheng and Jiang 2020; De Cicco et al. 2020; Hildebrand and Bergner 2021; Kushwaha et al. 2021; Mozafari et al. 2021; Nuruzzaman and Hussain 2020; Pillai and Sivathanu 2020). Eren (2021) confirmed that the users’ performance perceptions regarding the use of chatbots positively affected their customer satisfaction levels in the banking sector. This finding is in line with the expectancy violation theory, as individuals form expectations following their interactions with information systems (Chopra 2019; Neuburger et al. 2018).
The individuals’ social expectations from conversational technologies are especially pronounced when they incorporate cues of humanness (Adam et al. 2021; Pfeuffer et al. 2019), that are not present in traditional systems like websites, mobile applications, and databases (Belanche et al. 2021). The anthropomorphic features of AI dialogue systems make it easier for humans to connect with them (Adam et al. 2021; Becker et al. 2022; Forgas-Coll et al. 2022; Van Pinxteren et al. 2020).
In many cases, a number of quantitative researchers have investigated online users’ perceptions and attitudes toward these interactive technologies. Very often, they relied on valid measures that were tried and tested in academia. Some utilized the theory of reasoned action (Huang and Kao, 2021), the theory of planned behavior (Brachten et al. 2021; Ciechanowski et al. 2019), the behavioral reasoning theory (Lalicic and Weismayer 2021), the technology acceptance model (Kasilingam 2020) or the unified theory of acceptance and use of technology (Mostafa and Kasamani 5 2021), as they sought to investigate the individuals’ utilitarian motivations to use chatbot technologies to resolve their consumer issues. Others examined the users’ gratifications (Cheng and Jiang 2020; Rese et al. 2020), perceived enjoyment (De Cicco et al. 2020; Kushwaha et al. 2021; Rese et al. 2020), emotional factors (Crolic et al. 2021; Lou et al. 2021; Schepers et al. 2022; Wei et al. 2021), and/or intrinsic motivations (Jiménez-Barreto et al. 2021), to determine whether they were (or were not) affecting their intentions to use them.
This empirical study provides a snapshot of the online users’ perceptions about Chat Generative Pre-Trained Transformer (ChatGPT)’s responses to verbal queries, and sheds light on their dispositions to avail themselves from ChatGPT’s natural language processing.
It explores their performance expectations about their usefulness and their effort expectations related to the ease of use of these information technologies and investigates whether they are affected by colleagues or by other social influences to use such dialogue systems. Moreover, it examines their insights about the content quality, source trustworthiness as well as on the interactivity features of these text-generative AI models.
Generally, the results suggest that the research participants felt that these algorithms are easy to use. The findings indicate that they consider them to be useful too, specifically when the information they generate is trustworthy and dependable.
The respondents suggest that they are concerned about the quality and accuracy of the content that is featured in the AI chatbots’ answers. This contingent issue can have a negative effect on the use of the information that is created by online dialogue systems.
OpenAI’s ChatGPT is a case in point. Its app is freely available in many countries, via desktop and mobile technologies including iOS and Android. The company admits that its GPT-3.5 outputs may be inaccurate, untruthful, and misleading at times. It clarifies that its algorithm is not connected to the internet, and that it can occasionally produce incorrect answers (OpenAI, 2023a). It posits that GPT-3.5 has limited knowledge of the world and events after 2021 and may also occasionally produce harmful instructions or biased content.
OpenAI recommends checking whether its chatbot’s responses are accurate or not, and to let them know when and if it answers in an incorrect manner, by using their “Thumbs Down” button. They even declare that their ChatGPT’s Help Center can occasionally make up facts or “hallucinate” outputs (OpenAI, 2023a, OpenAI, 2023b).
OpenAI reports that its top notch ChatGPT Plus subscribers can access safer and more useful responses. In this case, users can avail themselves from a number of beta plugins and resources that can offer a wide range of capabilities including text-to-speech applications as well as web browsing features through Bing.
Yet again, OpenAI (2023b) indicates that its GPT-4 still has many known limitations that the company is working to address, such as “social biases and adversarial prompts” (at the time of writing this article). Evidently, works are still in progress at OpenAI.
The company needs to resolve these serious issues, considering that its Content Policy and Terms clearly stipulate that OpenAI’s consumers are the owners of the output that is created by ChatGPT. Hence, ChatGPT’s users have the right to reprint, sell, and merchandise the content that is generated for them through OpenAI’s platforms, regardless of whether the output (its response) was provided via a free or a paid plan.
Various commentators are increasingly raising awareness about the corporate digital responsibilities of those involved in the research, development and maintenance of such dialogue systems. A number of stakeholders, particularly the regulatory ones, are concerned on possible risks and perils arising from AI algorithms including interactive chatbots.
In many cases, they are warning that disruptive chatbots could disseminate misinformation, foster prejudice, bias and discrimination, raise privacy concerns, and could lead to the loss of jobs. Arguably, one has to bear in mind that, in many cases, many governments are outpaced by the proliferation of technological innovations (as their development happens before the enactment of legislation).
As a result, they tend to be reactive in the implementation of substantive regulatory interventions. This research reported that the development of ChatGPT has resulted in mixed reactions among different stakeholders in society, especially during the first months after its official launch.
At the moment, there are just a few jurisdictions that have formalized policies and governance frameworks that are meant to protect and safeguard individuals and entities from possible risks and dangers of AI technologies (Camilleri, 2023). Of course, voluntary principles and guidelines are a step in the right direction. However, policy makers are expected by various stakeholders to step-up their commitment by introducing quasi-regulations and legislation.
Currently, a number of technology conglomerates including Microsoft-backed OpenAI, Apple and IBM, among others, anticipated the governments’ regulations by joining forces in a non-profit organization entitled, “Partnership for AI” that aims to advance safe, responsible AI, that is rooted in open innovation.
In addition, IBM has also teamed up with Meta and other companies, startups, universities, research and government organizations, as well as non-profit foundations to form an “AI Alliance”, that is intended to foster innovations across all aspects of AI technology, applications and governance.
Suggested citation: Camilleri, M. A. (2024). Factors affecting performance expectancy and intentions to use ChatGPT: Using SmartPLS to advance an information technology acceptance framework. Technological Forecasting and Social Change, 201, https://doi.org/10.1016/j.techfore.2024.123247
This is an excerpt from one of my latest contributions on the use of responsive chatbots by service businesses. The content was adapted for this blogpost.
Suggested citation: Camilleri, M.A. & Troise, C. (2022). Live support by chatbots with artificial intelligence: A future research agenda. Service Business, https://doi.org/10.1007/s11628-022-00513-9
(Credit: Chatbots Magazine)
The benefits of using chatbots for online customer services
Frequently, consumers are engaging with chatbot systems without even knowing, as machines (rather than human agents) are responding to online queries (Li et al. 2021; Pantano and Pizzi 2020; Seering et al. 2018; Stoeckli et al. 2020). Whilst 13% of online consumer queries require human intervention (as they may involve complex queries and complaints), more than 87 % of online consumer queries are handled by chatbots (Ngai et al., 2021).
Several studies reported that there are many advantages of using conversational chatbots for customer services. Their functional benefits include increased convenience to customers, enhanced operational efficiencies, reduced labor costs, and time-saving opportunities.
Consumers are increasingly availing themselves of these interactive technologies to retrieve detailed information from their product recommendation systems and/or to request their assistance to help them resolve technical issues. Alternatively, they use them to scrutinize their personal data. Hence, in many cases, customers are willing to share their sensitive information in exchange for a better service.
Although, these interactive technologies are less engaging than human agents, they can possibly elicit more disclosures from consumers. They are in a position to process the consumers’ personal data and to compare it with prior knowledge, without any human instruction. Chatbots can learn in a proactive manner from new sources of information to enrich their database.
Whilst human customer service agents may usually handle complex queries including complaints, service chatbots can improve the handling of routine consumer queries. They are capable of interacting with online users in two-way communications (to a certain extent). Their interactions may result in significant effects on consumer trust, satisfaction, and repurchase intentions, as well as on positive word-of-mouth publicity.
Many researchers reported that consumers are intrigued to communicate with anthropomorphized technologies as they invoke social responses and norms of reciprocity. Such conversational agents are programed with certain cues, features and attributes that are normally associated with humans.
The findings from this review clearly indicate that individuals feel comfortable using chatbots that simulate human interactions, particularly with those that have enhanced anthropomorphic designs. Many authors noted that the more chatbots respond to users in a natural, humanlike way, the easier it is for the business to convert visitors into customers, particularly if they improve their online experiences. This research indicates that there is scope for businesses to use conversational technologies to personalize interactions with online users, to build better relationships with them, to enhance consumer satisfaction levels, to generate leads as well as sales conversions.
The costsof using chatbotsfor online customer services
Despite the latest advances in the delivery of electronic services, there are still individuals who hold negative perceptions and attitudes towards the use of interactive technologies. Although AI technologies have been specifically created to foster co-creation between the service provider and the customer,
There are a number of challenges (like authenticity issues, cognition challenges, affective issues, functionality issues and integration conflicts) that may result in a failed service interaction and in dissatisfied customers. There are consumers, particularly the older ones, who do not feel comfortable interacting with artificially intelligent technologies like chatbots, or who may not want to comply with their requests, for different reasons. For example, they could be wary about cyber-security issues and/or may simply refuse to engage in conversations with a robot.
A few commentators contended that consumers should be informed when they are interacting with a machine.In many cases, online users may not be aware that they are engaging with elaborate AI systems that use cues such as names, avatars, and typing indicators that are intended to mimic human traits.Many researchers pointed out that consumers may or may not want to be serviced by chatbots.
A number of researchers argued that some chatbots are still not capable of communicative behaviors that are intended to enhance relational outcomes. For the time being, there are chatbot technologies that are not programed to answer to all of their customers’ queries (if they do not recognize the keywords that are used by the customers), or may not be quick enough to deal with multiple questions at the same time. Therefore, the quality of their conversations may be limited. Such automated technologies may not always be in a position to engage in non-linear conversations, especially when they have to go back and forth on a topic with online users.
Theoretical and practical implications
This contribution confirms that recently there is a growing interest among academia as well as by practitioners on research that is focused on the use of chatbots that can improve the businesses’ customer-centric services. It clarifies that various academic researchers have often relied on different theories including on the expectancy theory, or on the expectancy violation theory, the human computer interaction theory/human machine communication theory, the social presence theory, and/or on the social response theory, among others.
Currently, there are limited publications that integrated well-established conceptual bases (like those featured in the literature review), or that presented discursive contributions on this topic. Moreover, there are just a few review articles that capture, scrutinize and interpret the findings from previous theoretical underpinnings, about the use of responsive chatbots in service business settings. Therefore, this systematic review paper addresses this knowledge gap in the academic literature.
It clearly differentiates itself from mainstream research as it scrutinizes and synthesizes the findings from recent, high impact articles on this topic. It clearly identifies the most popular articles from Scopus and Web of Science, and advances a definition about anthropomorphic chatbots, artificial intelligence chatbots (or AI chatbots), conversational chatbot agents (or conversational entities, conversational interfaces, conversational recommender systems or dialogue systems), customer experience with chatbots, chatbot customer service, customer satisfaction with chatbots, customer value (or the customers’ perceived value) of chatbots, and on service robots (robot advisors). It discusses about the different attributes of conversational chatbots and sheds light on the benefits and costs of using interactive technologies to respond to online users’ queries.
In sum, the findings from this research reveal that there is a business case for online service providers to utilize AI chatbots. These conversational technologies could offer technical support to consumers and prospects, on various aspects, in real time, round the clock. Hence, service businesses could be in a position to reduce their labor costs as they would require fewer human agents to respond to their customers. Moreover, the use of interactive chatbot technologies could improve the efficiency and responsiveness of service delivery. Businesses could utilize AI dialogue systems to enhance their customer-centric services and to improve online experiences. These service technologies can reduce the workload of human agents. The latter ones can dedicate their energies to resolve serious matters, including the handling of complaints and time-consuming cases.
On the other hand, this paper also discusses potential pitfalls. Currently, there are consumers who for some reason or another, are not comfortable interacting with automated chatbots. They may be reluctant to engage with advanced anthropomorphic systems that use avatars, even though, at times, they can mimic human communications relatively well. Such individuals may still appreciate a human presence to resolve their service issues. They may perceive that interactive service technologies are emotionless and lack a sense of empathy.
Presently, chatbots can only respond to questions, keywords and phrases that they were programed to answer. Although they are useful in solving basic queries, their interactions with consumers are still limited. Their dialogue systems require periodic maintenance. Unlike human agents they cannot engage in in-depth conversations or deal with multiple queries, particularly if they are expected to go back and forth on a topic.
Most probably, these technical issues will be dealt with over time, as more advanced chatbots will be entering the market in the foreseeable future. It is likely that these AI technologies would possess improved capabilities and will be programmed with up-to-date information, to better serve future customers, to exceed their expectations.
Limitations and future research avenues
This research suggests that this area of study is gaining traction in academic circles, particularly in the last few years. In fact, it clarifies that there were four hundred twenty-one 421 publications on chatbots in business-related journals, up to December 2021. Four hundred fifteen (415) of them were published in the last 5 years.
The systematic analysis that was presented in this research was focused on “chatbot(s)” or “chatterbot(s)”. Other academics may refer to them by using different synonyms like “artificial conversational entity (entities)”, “bot(s)”, “conversational avatar(s)”, “conversational interface agent”, “interactive agent(s)”, “talkbot(s)”, “virtual agent(s)”, and/or “virtual assistant(s)”, among others. Therefore, future researchers may also consider using these keywords when they are other exploring the academic and nonacademic literature on conversational chatbots that are being used for customer-centric services.
Nevertheless, this bibliographic study has identified some of the most popular research areas relating to the use of responsive chatbots in online customer service settings. The findings confirmed that many authors are focusing on the chatbots’ anthropomorphic designs, AI capabilities and on their dialogue systems. This research suggests that there are still knowledge gaps in the academic literature. The following table clearly specifies that there are untapped opportunities for further empirical research in this promising field of study.
The full article is forthcoming. A prepublication version will be available through Researchgate.
Substantia Mea 
You must be logged in to post a comment.