Artificial Intelligence Agents Research Articles

Empathic accuracy in artificial intelligence service recovery

PropósitoLas agentes de inteligencia artificial (IA) se aplican cada vez más en el sector del turismo y la hostelería. Sin embargo, los fallos de los servicios de IA son inevitables. Por lo tanto, la recuperación de servicios de IA merece una investigación empírica. Esta investigación explora cómo la precisión empática de la IA afecta a la satisfacción de los clientes con la recuperación del servicio de IA.Diseño/Metodología/EnfoqueSe presentó un modelo de mediación moderado para describir el efecto de la precisión empática en la satisfacción del cliente mediante cuatro experimentos basados en escenarios.HallazgosLos resultados revelan el impacto positivo de la precisión empática de la IA en la satisfacción del cliente y los efectos mediadores de la agencia percibida y la experiencia percibida. Además, el antropomorfismo modera el efecto de la precisión empática.OriginalidadEste artículo amplía los estudios sobre los servicios de IA al investigar el papel de la precisión empática en la satisfacción del cliente. Los resultados aportan un punto de vista teórico novedoso sobre la retención de clientes tras el fallo de un servicio de IA.

Scientometric analysis on the use of ChatGPT, artificial intelligence, or intelligent conversational agent in the role of medical training

IntroductionIn recent times, there has been a noticeable surge in the usage of artificial intelligence, including ChatGPT and other types, in the field of health sciences education. In this regard, an exploratory bibliometric study was carried out to examine the utilization of smart conversational agents, ChatGPT, and artificial intelligence bots in medical education. MethodsA retrospective, observational, cross-sectional bibliometric analysis was employed to assess the scientific publications listed in Scopus. This study was conducted on March 11, 2023 in search for information in Scopus. A total of 220 relevant documents were identified that were available in the Scopus database during the period between 2017 and 2022. Elsevier's SciVal software was used. Subsequently, statistical tables and graphs were prepared for presentation in Bibliometrix software. ResultsAmong the authors, Timothy W. Bickmore, from the United States, has the highest number of publications (10) and citations received (172), and an h-index of 45, suggesting a significant influence in the field of study. The subcategory with the highest academic output is Health Informatics with 133 publications, while Geriatrics and Gerontology has the least with only 3. Most of the analyzed publications (44.2%) originated from collaborations within the same country. Notably, the Swiss Federal Institute of Technology Zurich and Imperial College London stood out with 12 publications each that received over 200 citations indicating their significant impact on their respective fields. Despite having the highest number of academic publications (15), Brazil had a relatively low field-weighted citation impact (0.64) and received the lowest number of citations (81). A clustering analysis was performed on a sample of 10 concepts using 2 dimensions. The results indicated that all terms were part of the same cluster. Notably, the terms 'conversational agents', 'chatbots', 'conversational agent', and 'chatbot' were closely related. ConclusionsIt was found that the American Bickmore, Timothy W., led the top-10 researchers, and that the Health Informatics subject area was the most predominant. However, Brazil and Germany were the leading countries in terms of research output that was mainly published in high impact journals (Q1).

Building Socially Intelligent AI Systems: Evidence from the Trust Game Using Artificial Agents with Deep Learning

The trust game, a simple two-player economic exchange, is extensively used as an experimental measure for trust and trustworthiness of individuals. We construct deep neural network–based artificial intelligence (AI) agents to participate a series of experiments based upon the trust game. These artificial agents are trained by playing with one another repeatedly without any prior knowledge, assumption, or data regarding human behaviors. We find that, under certain conditions, AI agents produce actions that are qualitatively similar to decisions of human subjects reported in the trust game literature. Factors that influence the emergence and levels of cooperation by artificial agents in the game are further explored. This study offers evidence that AI agents can develop trusting and cooperative behaviors purely from an interactive trial-and-error learning process. It constitutes a first step to build multiagent-based decision support systems in which interacting artificial agents are capable of leveraging social intelligence to achieve better outcomes collectively. This paper was accepted by Yan Chen, behavioral economics and decision analysis. Funding: Y. (D.) Wu extends her gratitude for the financial support provided through the RSCA Seed [Grant 22-RSG-01-004] from the San Jose State University. Supplemental Material: Data are available at https://doi.org/10.1287/mnsc.2023.4782 .

Modeling Nonusers’ Behavioral Intention towards Mobile Chatbot Adoption: An Extension of the UTAUT2 Model with Mobile Service Quality Determinants

Artificial intelligence agents (chatbots), which are programs to communicate with users primarily in customer service contexts, are an alternative interaction channel supporting businesses in the digital environment and vital components in customer service. The present empirical paper, which is aimed at identifying and discussing the factors motivating nonusers to adopt the specific technology in mobile contexts, proposes a comprehensive conceptual model, which combines the UTAUT 2 behavioral theory with variables of mobile service quality contexts, such as information quality, privacy concerns, interface, and equipment, as well as trust and mobility factors. Data analysis, based on the partial least squares structural equation modeling (PLS-SEM) statistical method, revealed that performance expectancy, facilitating factors, hedonic motivation, mobility, trust, and service quality positively affect nonusers’ behavioral intention to adopt chatbots. In addition, equipment, interface, and trust have a significant impact on users’ trust in the context of mobile chatbots. Personal data privacy issues also have a negative effect on trust, in contrast to effort expectancy, which positively affects performance expectancy. As mobile service quality factors have not been investigated before in the context of chatbots, the findings of the present research are expected to provide useful insights both to academia and the business industry.

Examining ChatGPT Performance on USMLE Sample Items and Implications for Assessment.

In late 2022 and early 2023, reports that ChatGPT could pass the United States Medical Licensing Examination (USMLE) generated considerable excitement, and media response suggested ChatGPT has credible medical knowledge. This report analyzes the extent to which an artificial intelligence (AI) agent's performance on these sample items can generalize to performance on an actual USMLE examination and an illustration is given using ChatGPT. As with earlier investigations, analyses were based on publicly available USMLE sample items. Each item was submitted to ChatGPT (version 3.5) 3 times to evaluate stability. Responses were scored following rules that match operational practice, and a preliminary analysis explored the characteristics of items that ChatGPT answered correctly. The study was conducted between February and March 2023. For the full sample of items, ChatGPT scored above 60% correct except for one replication for Step 3. Response success varied across replications for 76 items (20%). There was a modest correspondence with item difficulty wherein ChatGPT was more likely to respond correctly to items found easier by examinees. ChatGPT performed significantly worse ( P < .001) on items relating to practice-based learning. Achieving 60% accuracy is an approximate indicator of meeting the passing standard, requiring statistical adjustments for comparison. Hence, this assessment can only suggest consistency with the passing standards for Steps 1 and 2 Clinical Knowledge, with further limitations in extrapolating this inference to Step 3. These limitations are due to variances in item difficulty and exclusion of the simulation component of Step 3 from the evaluation-limitations that would apply to any AI system evaluated on the Step 3 sample items. It is crucial to note that responses from large language models exhibit notable variations when faced with repeated inquiries, underscoring the need for expert validation to ensure their utility as a learning tool.

Artificial intelligence applications in aviation accident classification: A preliminary exploratory study

Artificial intelligence (AI) applications are already integrated and expected to expand within the aviation sociotechnical ecosystem. Machine Learning (ML) models are capable and could excel on accident report classification tasks within the aviation context, assuming adequate and proper training. Human subject matter experts (SMEs) and the open access AI ChatGPT reanalyzed the TransAsia Airways flight GE235, Air France AF447, and Helios Airways Flight HCY552 accidents using the Human Factors Analysis and Classification System (HFACS) to investigate whether the AI agent is capable of conducting aviation accident analysis and comparing raters’ results, of verifying AI’s learning abilities, and, finally, investigating whether AI analysis could expand in other systems-based models. The results suggested that: (a) AI/ML applications could be an efficient tool for learning organizations’ continuous learning, growth, and improvement by creating new knowledge from previous failures, (b) the proper prompts and well-established frameworks could optimize the agreement between human SMEs and AI in accident analysis, (c) the ChatGPT performed better on linear models (i.e., Bowtie) compared to complex systemic accident causation models (e.g., HFACS), (d) the ChatGPT could not conduct high- and low-level analysis simultaneously, and, (e) a multidisciplinary team including human factors, IT, and accident investigators experts is essential in developing and overseeing the W-shaped learning assurance process.

Mobile robot navigation based on Deep Reinforcement Learning: A brief review

Navigation capacity is a key attribute of robot technology and the foundation for achieving other advanced behaviours. Compared to traditional navigation technology, applying Deep Reinforcement Learning (DRL) to artificial intelligence agents to achieve mobile robot navigation function is currently the academic focus. DRL is based on an end-to-end approach, transforming high-dimensional and continuous inputs into optimal policy to guide mobile robots, forming an advanced perceptual control system. In this article, DRL is first compared with traditional navigation technology and SLAM, and its application advantages are elucidated. Then, the basic background and classic algorithm models of standard reinforcement learning and DRL are systematically elaborated. Finally, the application of DRL in different application scenarios and research fields is introduced.

Part One (Describing Criminality): The Society Agent Simulating (SAS) &amp; Usage of Oral and Dental Diseases to Predict Criminals Earlier

Agent Based Modeling and Simulation &amp; Usage of Oral and Dental Diseases to predict criminals &amp; crimes: in part one (this paper) of our research description we only try to describe the knowledge of Cognitive Science, Agent-Based Simulation, Social Simulation, Artificial Intelligence and Multi Agent Systems. In the other hand in the second part (next paper) we will show how we can use the oral &amp; dental diseases as a most important data for the Agent-Based Social Simulation, to early predict the criminals and crimes. In social simulating to create social behavioral modeling we need to apply the cognitive abilities. The cognitive ability is the most important property for the social agent. But the agent architecture is not included this ability. The agent with the cognitive ability will be able to improve the models of the behaviors in the society. This ability not only will be useful in the area of the society security and immigrant’s behavior in the target society, but also it is very useful in business and management fields. In this paper we are going to indicate different fields that can be used to predict the behavior of human. To know the different behaviors can make us to be able to predict criminals earlier by applying the comparison target’s internal harmony of life with the external harmony of world. So we will be able to predict the criminals before doing the crimes. It will help us to have secure life. Besides it will be useful in any other fields of life such as business, banking, political aspects, human resources and etc.

Coloring and fusing architectural sketches by combining a Y‐shaped generative adversarial network and a denoising diffusion implicit model

AbstractSketch‐based architectural design tasks require considerable time and effort. However, developing an artificial intelligence agent to color and fuse architectural sketches is challenging due to the imaginative nature of the task and technical limitations. Here, we first introduce Y‐shaped generative adversarial networks (GANs), which can color sketches on the basis of temporarily specified reference images by embedding an attention mechanism model that can match spatial and color information and by maintaining a type of triangular equilibrium state. Then, we develop practicable methods for coloring and fusing the given sketches simultaneously using a combination of our Y‐shaped GANs and the denoising diffusion implicit model, yielding results with high quality and veracity. A novel interpolation process is developed as a necessary trick for realizing the fusion function. The testing results indicate that this method works better than the existing methods for the tasks of coloring and fusing architectural sketches.

Humor type and service context shape AI service recovery

While artificial intelligence (AI) service failure is inevitable in tourism and hospitality, scholarly attention to it is limited. This study investigated how the type of humor (self-deprecating or self-enhancing) that AI agents adopt interacted with the service context (hedonic-dominant or utilitarian-dominant) in influencing customers' intentions to continue using AI devices. Findings indicate that in hedonic-dominant service contexts, self-deprecating and self-enhancing humor increase continuous usage intention; in utilitarian-dominant service contexts, only self-deprecating humor boosts continuous usage intention. Results also reveal the mediating roles of positive emotion and inferred negative motives along with the moderating effect of anthropomorphism. This study enriches theories on AI service recovery and offers tourism firms suggestions on how to use humor to resolve AI service failures.

The impact of AI identity disclosure on consumer unethical behavior: A social judgment perspective

Implementing artificial intelligence (AI) technology has resulted in various advantages but also the emergence of unethical conduct among consumers. This article examines the impact of the disclosure of AI identities on consumer unethical behavior through pilot experiments and three formal experiments. The results indicate that the disclosure of AI agents leads to a higher propensity of consumers, demonstrating unethical behavior compared to when AI agents are undisclosed. Perceived social judgment plays a crucial role in mediating this effect. Additionally, this research ruled out other potential explanatory mechanisms, such as warmth, empathy, and competence. Furthermore, consumers' motive for impression management acts as a moderator. These findings provide valuable insights into the impact of AI identity disclosure on consumer behavior and offer guidance for businesses to mitigate unethical consumer behavior.

Video Turing Test: A first step towards human‐level AI

AbstractThe development of artificial intelligence (AI) agents capable of human‐level understanding of video content and conducting conversations with humans on this basis is a promising application that people expect. However, this is a challenging task that requires the holistic integration of multimodal information with temporal dependencies and reasoning, as well as social and physical commonsense. In addition, the development of appropriate systematic evaluation methods is essential. In this context, we introduce the Video Turing Test (VTT), a blind test used to evaluate human‐likeness in terms of video comprehension ability. Moreover, we propose Vincent as a video understanding AI. We explain the configuration of VTT, the architecture of Vincent to prepare for VTT and the proposed evaluation methods for video comprehension. We also estimate the current intelligence level of AI based on our results and discuss future research directions.

How gamifying AI shapes customer motivation, engagement, and purchase behavior

AbstractRecently, firms started to gamify conversational artificial intelligence (AI) agents, such as chatbots, to improve purchase outcomes. This article explores strategies for incorporating gamification into AI systems by investigating the impact of utilitarian and hedonic motivations facilitated by gamified chatbots on various dimensions of customer engagement (cognitive, emotional, and behavioral), as well as the subsequent effects of these dimensions on customers' purchase behavior. By conducting one cross‐sectional and two experimental studies involving real interactions with gamified chatbots, this research identifies two crucial paths that warrant attention: an optimal path from hedonic motivation to behavioral engagement, resulting in enhanced purchase, and a detrimental path from utilitarian motivation to emotional engagement, which reduces purchase. Furthermore, the research compares the effects of two types of gamified chatbots and reveals that a game‐of‐chance‐based chatbot, as opposed to a knowledge‐sharing gamified chatbot, aligns with the optimal path, leading to higher purchasing while at the same time avoiding that customers feel obligated to play the game. Based on these findings, the article provides actionable insights for eliciting favorable psychological and behavioral responses through gamified AI interactions.

Effects of contemporary technologies, such as blockchain and artificial intelligence (AI) in enhancing consumers’ trustworthiness of online reviews

ABSTRACT The research note discusses the potential effects of contemporary technologies, such as blockchain and artificial intelligence (AI), in enhancing consumers’ trustworthiness of online reviews in the tourism and hospitality context. The study postulates that three characteristics of blockchain – decentralization, immutability, and timestamping – ensure transparency in the online review process. Using a socio-technical systems approach, the current study provides insights into how blockchain technology, AI agents, and human factors interact to facilitate such transparency and thereby enhance consumers’ trustworthiness of online reviews for their decision-making. Therefore, the study proposes that perceived transparency mediates the relationship between these three blockchain characteristics and the trustworthiness of online reviews. In proposing an investigation of the characteristics of blockchain and its interaction with AI in online reviews, the study highlights that researchers and practitioners pay attention to a specific characteristic of contemporary technologies to actualize potential benefits.

Examining the Impact of Assistive Technology on Psychological Health, Family Education, and Curriculum Research in Japan: Insights from Artificial Intelligence.

This study aims to analyze the effect of psychological health based on artificial intelligence agent technology on the implementation effect of Japanese family education. By combining mobile agent technology and education thought, the system structure and working mechanism of the education support system of agents are studied to build personalized support for the family education system based on mobile agents. A total of 320 Japanese middle school students were randomly divided into an experimental group and a control group, with 160 cases in each group. The control group received traditional family health education, while the experimental group received mental health education based on the Agent Technology family education system. The basic information and mental health scores of the two groups of students were compared. The results showed that there were no remarkable differences in the number of male and female cases, weight, height, average age, grade, home address, or family situation between groups (p > 0.05). The psychological health level of the experimental group was considerably superior to that of the control group regarding obsessional symptoms, interpersonal tension and sensitivity, depression, anxiety, learning pressure, maladaptation, emotional imbalance, and psychological imbalance (p < 0.05). In summary, compared with traditional family education, family education of the mental health education system based on agent technology can better improve the level of middle school students' mental health, which can improve student forced symptoms, interpersonal tension and sensitivity, depression, anxiety, learning pressure, maladjustment, emotional imbalance, psychological imbalance, and many other psychological states. Furthermore, personalized support for family education systems based on mobile agents has the advantages of autonomy, responsiveness, initiative, and mobility, which provides a new idea for family education.

Artificial intelligence and conversational agent evolution – a cautionary tale of the benefits and pitfalls of advanced technology in education, academic research, and practice

PurposeThe purpose of this paper is to demonstrate the advancements in artificial intelligence (AI) and conversational agents, emphasizing their potential benefits while also highlighting the need for vigilant monitoring to prevent unethical applications.Design/methodology/approachAs AI becomes more prevalent in academia and research, it is crucial to explore ways to ensure ethical usage of the technology and to identify potentially unethical usage. This manuscript uses a popular AI chatbot to write the introduction and parts of the body of a manuscript discussing conversational agents, the ethical usage of chatbots and ethical concerns for academic researchers.FindingsThe authors reveal which sections were written entirely by the AI using a conversational agent. This serves as a cautionary tale highlighting the importance of ethical considerations for researchers and students when using AI and how educators must be prepared for the increasing prevalence of AI in the academy and industry. Measures to mitigate potential unethical use of this evolving technology are also discussed in the manuscript.Originality/valueAs conversational agents and chatbots increase in society, it is crucial to understand how they will impact the community and how we can live with technology instead of fighting against it.

More than just a chat: a taxonomy of consumers’ relationships with conversational AI agents and their well-being implications

PurposeThis paper aims to study the role of self-concept in consumer relationships with anthropomorphised conversational artificially intelligent (AI) agents. First, the authors investigate how the self-congruence between consumer self-concept and AI and the integration of the conversational AI agent into consumer self-concept might influence such relationships. Second, the authors examine whether these links with self-concept have implications for mental well-being.Design/methodology/approachThis study conducted in-depth interviews with 20 consumers who regularly use popular conversational AI agents for functional or emotional tasks. Based on a thematic analysis and an ideal-type analysis, this study derived a taxonomy of consumer–AI relationships, with self-congruence and self–AI integration as the two axes.FindingsThe findings unveil four different relationships that consumers forge with their conversational AI agents, which differ in self-congruence and self–AI integration. Both dimensions are prominent in replacement and committed relationships, where consumers rely on conversational AI agents for companionship and emotional tasks such as personal growth or as a means for overcoming past traumas. These two relationships carry well-being risks in terms of changing expectations that consumers seek to fulfil in human-to-human relationships. Conversely, in the functional relationship, the conversational AI agents are viewed as an important part of one’s professional performance; however, consumers maintain a low sense of self-congruence and distinguish themselves from the agent, also because of the fear of losing their sense of uniqueness and autonomy. Consumers in aspiring relationships rely on their agents for companionship to remedy social exclusion and loneliness, but feel this is prevented because of the agents’ technical limitations.Research limitations/implicationsAlthough this study provides insights into the dynamics of consumer relationships with conversational AI agents, it comes with limitations. The sample of this study included users of conversational AI agents such as Siri, Google Assistant and Replika. However, future studies should also investigate other agents, such as ChatGPT. Moreover, the self-related processes studied here could be compared across public and private contexts. There is also a need to examine such complex relationships with longitudinal studies. Moreover, future research should explore how consumers’ self-concept could be negatively affected if the support provided by AI is withdrawn. Finally, this study reveals that in some cases, consumers are changing their expectations related to human-to-human relationships based on their interactions with conversational AI agents.Practical implicationsThis study enables practitioners to identify specific anthropomorphic cues that can support the development of different types of consumer–AI relationships and to consider their consequences across a range of well-being aspects.Originality/valueThis research equips marketing scholars with a novel understanding of the role of self-concept in the relationships that consumers forge with popular conversational AI agents and the associated well-being implications.

Trust in Artificial Intelligent Agent while Completing a Procedural Construction Task

The use of AI-enabled recommender systems in construction activities has the potential to improve worker performance and reduce errors; however, the accuracy of such systems in providing effective suggestions is dependent on the quality of their training data. A within-subjects experimental study was conducted using a simulated recommender system for installation tasks to investigate the effect of system reliability and construction task complexity on worker trust, workload, and performance. Results indicate that overall trust in the AI agent was higher for the highly reliable condition but remained consistent across various levels of task complexity. The workload was found to be higher for low reliability and complex conditions, and the effect of reliability on performance was influenced by task complexity. These findings offer insights for designing recommender systems to support construction workers in completing procedural tasks.

On the Use of Asset Administration Shell for Modeling and Deploying Production Scheduling Agents within a Multi-Agent System

Industry 4.0 (I4.0) aims at achieving the interconnectivity of multiple industrial assets from different hierarchical layers within a manufacturing environment. The Asset Administration Shell (AAS) is a pilar component of I4.0 for the digital representation of assets and can be applied in both physical and digital assets, such as enterprise software, artificial intelligence (AI) agents, and databases. Multi-agent systems (MASs), in particular, are useful in the decentralized optimization of complex problems and applicable in various planning or scheduling scenarios that require the system’s ability to adapt to any given problem by using different optimization methods. In order to achieve this, a universal model for the agent’s information, communication, and behaviors should be provided in a way that is interoperable with the rest of the I4.0 assets and agents. To address these challenges, this work proposes an AAS-based information model for the description of scheduling agents. It allows multiple AI methods for scheduling, such as heuristics, mathematical programming, and deep reinforcement learning, to be encapsulated within a single agent, making it adjustable to different production scenarios. The software implementation of the proposed architecture aims to provide granularity in the deployment of scheduling agents which utilize the underlying AAS metamodel. The agent was implemented using the SARL agent-oriented programming (AOP) language and deployed in an open-source MAS platform. The system evaluation in a real-life bicycle production scenario indicated the agent’s ability to adapt and provide fast and accurate scheduling results.

More Than a Bot? The Impact of Disclosing Human Involvement on Customer Interactions with Hybrid Service Agents

To leverage the complementary strengths of humans and artificial intelligence (AI) in online service encounters, firms have begun to use hybrid service agents: combinations of AI agents (e.g., chatbots) and human agents (e.g., service employees) behind a single interface. However, it is unclear whether firms should be transparent about behind-the-scenes employees working in tandem with an AI-based chatbot to serve customers. Against this backdrop, we investigated the impact of human involvement disclosure on customer interactions with hybrid service agents. Our findings suggest that disclosing human involvement before or during an interaction with the hybrid service agent leads customers to adopt a more human-oriented communication style. This effect is driven by impression management concerns that are activated when customers become aware of humans working in tandem with the chatbot. The more human-oriented communication style ultimately increases employee workload because fewer customer requests can be handled automatically by the chatbot and must be delegated to a human. These findings provide novel insights into how and why disclosing human involvement affects customer communication behavior, reveal its negative consequences for employees working in tandem with a chatbot, and highlight the potential costs and benefits of providing transparency in customer–hybrid service agent interactions.