Advances In Artificial Intelligence Research Articles

From extracting insights from large-scale, multimodal data to prevention and support, there is growing interest in the applications and implications of recent advances in Artificial Intelligence (AI) within the fields of addiction, substance use and mental health, which we refer to as ASUM. However, due to the absence of a structured mapping of AI for ASUM, it remains unclear how this interest is translated into concrete research results. This paper addresses this gap by conducting a bibliometric analysis of AI for ASUM, exploring: (i) the scale of ASUM-related research (number of publications, authors, institutions and countries); (ii) the evolution of ASUM's research productivity over time, both in absolute terms and relative to its parent disciplines; (iii) the key topics within ASUM and their interrelations. Results, supplemented by a comparison of similar fields, show that, while ASUM is an emerging and rapidly expanding domain (with a 25-fold increase in research output since 2012, attracting growing attention relative to parent disciplines as well as appearing to rely on applying more advanced AI methods than related fields), it remains largely fragmented through a dispersed group of infrequent contributors. An integration of the findings suggests two dominant trajectories through which AI for ASUM is currently being realised: as AI-driven analytic support and as innovative research and therapeutic methods (e.g., virtual reality, chatbots). The paper concludes by situating AI for ASUM as an emerging scientific field, outlining the scientific and practical challenges and opportunities that are likely to arise, and high-potential research areas open for exploration.

Advent of Artificial Intelligence in Patient Information Leaflets: a stakeholders' survey.

Artificial intelligence (AI) is rapidly transforming the delivery of kidney care through predictive analytics, machine learning, deep learning, and generative AI technologies. To meet this challenge, the American Society of Nephrology convened an AI Workgroup to provide a framework for the responsible use of AI in nephrology. The group outlines foundational principles to guide AI development: prioritizing patient benefit, ensuring clinician oversight, and advancing innovation in high-burden disease areas. Its set of foundational assumptions are grounded in the physician always being in the loop and an overarching goal to benefit patients with kidney disease. This review provides an overview of the clinical uses of AI in nephrology and offers practical guidance for nephrologists seeking to incorporate AI into CKD and AKI management, dialysis, and transplantation care. It also highlights key challenges-such as data quality, equity, transparency, and clinical integration-that must be addressed to ensure the responsible and effective implementation of AI in kidney care.

There are an increasing number of identified benefits for employing generative artificial intelligence (GenAI) in the presentation of health information. For patients, GenAI is particularly appealing as it presents information in a way that is digestible and understandable to a broad audience. This includes tools such as ChatGPT, Gemini, Claude, Deepseek, and Copilot. However, the practice runs the risk of perpetuating misinformation and presenting potentially hazardous advice. With the rising development of antimicrobial resistance in urinary tract infections (UTIs), this disorder presents an area where it is important to ‘get the message right’. In some cases, a GenAI response to a health query can be considerably influenced by the addition of a single word to the user’s entered prompt. In particular, this study identified particular words of concern related to UTI treatments, such as ‘homeopathy’, ‘crystals’ and ‘star sign’. When a single pseudoscientific term (usually only a single word) was entered, the response tended to emphasise this, returning misinformation. This article provides insights into where concern should be taken if patients mention using GenAI for their health advice. The outcome is that health professionals should be specifically aware of the tendency for GenAI to emphasise pseudoscientific concepts if prompted, and its tendency to present them as effective interventions for UTIs. When counselling patients, a discussion of this concept would be necessary to clarify the limitations of this technology in health education.

Aim. The work aimed to identify the specific features of cross-border business development in the field of generative artificial intelligence (AI) and develop recommendations for the formation of promising strategies for Russian companies and the Government of the Russian Federation (RF). Objectives. The work seeks to identify the characteristic features of cross-border business and conduct a foresight session on its development prospects in the field of generative AI; to assess the main development trends of international business in the field of generative AI; to determine the specific features of Russian business development in the field of generative AI; to propose conceptual directions for a strategy to support cross-border business in the field of generative AI in Russia. Methods. The research methods include socioeconomic foresight using the “Four Worlds” technique, an expert survey, a literature review, conceptual and statistical analysis, and data extrapolation. Results. The work presents a scientific definition of cross-border business in the field of generative AI and characterizes organizational models for businesses using generative AI, commonly used in the modern context. The reasons for the steady transnationalization of generative AI business are substantiated, and its characteristics are highlighted. Approaches to and challenges in assessing the economic prospects for the development of international generative AI business are examined in detail, and the results of a foresight session on cross-border business development in generative AI are presented. The work revealed the potential of cross-border generative AI business originating in Russia, and identified the barriers to its development and opportunities for overcoming them, which are traced in the intensification of strategic interaction between the state and business. Conclusions. The most probable scenario for the development of cross-border online business in the field of generative AI is a combination of technological integration and competition among major players. For Russian companies and the Russian government, the key challenge in this regard is overcoming external obstacles and creating conditions for integrating own technologies into global value chains.

Purpose This study aims to explore the evolving research landscape of generative artificial intelligence (Gen-AI) by conducting a comprehensive bibliometric analysis. It seeks to identify prevailing themes, influential contributors and underexplored areas within the Gen-AI domain. Design/methodology/approach A bibliometric approach was adopted using data extracted from the Scopus database, focusing on publications from 2019–2024. The study used the Bibliometrix R package to analyze keyword trends, publication output, citation impact, country-wise productivity and thematic evolution. Findings The results reveal a significant rise in academic interest in Gen-AI, particularly since the release of models like ChatGPT. Key research areas include natural language processing, AI ethics, chatbots and education. However, several gaps persist – particularly in ethical use, bias mitigation and governance frameworks – suggesting the need for more focused and interdisciplinary research. Research limitations/implications The study is limited to publications indexed in the Scopus database and written in English. Valuable insights from other databases or non-English literature may have been excluded. The rapidly evolving nature of the field may also outpace the analysis over time. Originality/value This research provides a timely and data-driven overview of Gen-AI scholarship, offering actionable insights for researchers, practitioners and policymakers. It also highlights critical areas requiring further inquiry, contributing to a more balanced and ethically grounded advancement of generative technologies.

Purpose This study aims to analyze whether artificial intelligence (AI) (vs human) as a designer and offering users’ free choices for product customization (vs limited predefined choice sets) influence product and site attitude, and purchase intention. In the free choice condition, participants receive an open-ended textbox to write their preferences, akin to current Generative AI (GenAI) technologies. Design/methodology/approach A 2 (Designer: AI vs human) × 3 (Customization: low choice set vs medium choice set vs free choice) between-subject (N = 570) online experiment was conducted using an interface mimicking an e-commerce clothing company site. Analyses of covariance and mediation analyses were used to test the hypotheses. Findings Results revealed no difference (or negligible effect size) between AI and human designers for any outcome variables. However, regardless of the designer, the free choice condition led to more positive values for all outcome variables. The mediation analyses showed that this occurs because having free choice increases consumer agency. Originality/value GenAI advancements allow product customization based on user queries (vs a predefined limited choice set). However, an underexplored area is whether users would accept AI as an apparel designer and whether providing users with free choice for customization via prompting (a unique affordance of GenAI) will make a difference.

With the rapid advancement of technology, generative artificial intelligence is gradually permeating the field of education. Functional experiment teaching constitutes a vital component of medical education. However, traditional mod-els struggle to meet the demands of personalized learning and innovation ca-pacity cultivation due to limitations in experimental equipment and resources, rigid teaching methodologies, and untimely resource updates. This paper delves into the reformative concepts and practical pathways for applying generative artificial intelligence in functional experiment teaching, analyzing its notable advantages in enhancing teaching effectiveness and fostering students' innova-tive abilities. Simultaneously, it reflects on potential challenges during imple-mentation, aiming to provide valuable insights for the innovative development of functional experiment teaching.

ABSTRACT As generative artificial intelligence (Gen AI) rapidly transforms educational practices and reshapes workforce skill demands, vocational education faces unique challenges in adapting to these changes. Understanding vocational teachers’ perceptions and adoption of Gen AI is thus essential for ensuring effective integration of emerging technologies into teaching. The current study explored the factors influencing vocational teachers’ adoption of Gen AI in China, with a focus on disciplinary and regional differences. Structural equation modeling (SEM) was conducted on survey data from 1028 teachers. Results indicated that perceived usefulness, perceived easy of use, social influence, and facilitating conditions significantly shaped behavioral intention and actual usage. Multi-group SEM analysis revealed disciplinary differences: Humanities and Social Sciences teachers were more influenced by social norms, whereas Natural Sciences teachers were more responsive to perceived usefulness and easy of use. No significant regional differences were observed. These findings provide valuable insights for policymakers and institutions aiming to support diverse teacher groups in integrating Gen AI into vocational education.

Social robotics is a rapidly advancing field dedicated to the development of embodied artificial agents capable of social interaction with humans. These systems are deployed across domains such as health care, education, service, and entertainment-contexts that demand nuanced social competence. Yet, the social dimension of social robotics remains insufficiently conceptualized and empirically grounded. Many companies have failed as their robots struggle to sustain meaningful, long-term engagement with users. Understanding human responses to these agents requires robust psychological frameworks. While prior work has emphasized emotion expression and affective cues, human social interaction is shaped by broader constructs, including individual goals and roles, self-presentation, and culture. Generative artificial intelligence is reshaping human-robot interaction but has yet to resolve foundational challenges in social engagement. Addressing these gaps necessitates deeper integration of psychological theory, methodology, and data. A sustained dialogue between psychology and robotics holds promise not only for advancing socially adept machines but also for enriching psychological science itself.

Abstract The emergence of artificial intelligence (AI) innovations like ChatGPT presents new opportunities and challenges for world languages (WL) education. WL teacher education programs must prepare preservice teachers with AI literacy to help them effectively integrate these technologies into teaching. This multiple case study—part of our on‐going self‐study of teacher educator practice—investigated how AI literacy can be leveraged as a core practice in a WL teacher licensure program. Drawing on pre‐surveys, course artifacts, structured reflections, and interviews, the study explored how three teacher candidates (TCs) engaged with Generative AI (chatbots) in the instructional activity of lesson planning and developed emergent forms of AI literacy. Participants demonstrated varying levels of development of AI literacy across four domains: technological proficiency, pedagogical compatibility, professional work, and ethical use. They developed critical stances toward AI, which were shaped by their evolving professional identities. This study contributes to growing conversations about AI in teacher education by showing the potential of the AI literacy core practice as a scaffolded, reflective approach to building AI competencies. It also underscores the importance of centering TC's professional identity development in AI integration while providing support for prompt design, noninstructional use of AI, and facilitating conversations about responsible AI use with students.

The study explores the pedagogical and technological potential of the online platform Canva in the context of generative artificial intelligence (AI). The article focuses on how the integration of AI-powered tools within Canva’s Magic Studio – namely Magic Write, Text-to-Image, and Canva Code – transforms the process of creating educational materials and enhances teachers’ digital and creative competencies. The research aim is to analyze the educational value of Canva as a multifunctional AI ecosystem that supports teachers in content creation, visualization of complex concepts, and development of interactive learning tools without programming skills. Methodologically, the study employs analytical and experimental approaches to demonstrate the implementation of Canva’s AI features in the design of text-based, visual, and interactive educational resources. The results indicate that Magic Write significantly accelerates the preparation of lesson materials, Text-to-Image facilitates the generation of customized illustrations, and Canva Code enables interactive components that promote engagement and inclusivity in the learning environment. The use of Canva in education contributes to the realization of STEM and STEAM principles, encouraging interdisciplinary learning, creativity, and problem-solving. The findings confirm that Canva serves not merely as a design platform but as an intelligent, human-centered system that fosters a new culture of educational design adapted to the needs of the 21st century. Prospective directions for future research involve addressing ethical, copyright, and legal aspects of using generative AI in educational content creation.

Abstract A multi-agent artificial intelligence (AI) model is developed to automate the discovery of new metallic alloys, integrating multimodal data and external knowledge, including insights from physics via atomistic simulations. The system consists of (a) large language models (LLMs) for tasks such as reasoning and planning, (b) AI agents with distinct roles collaborating dynamically, and (c) a newly developed graph neural network (GNN) model for rapid retrieval of physical properties. We chose the ternary NbMoTa body-centered-cubic alloy as our model system and developed the GNN to predict two fundamental materials properties: the Peierls barrier and the solute/screw dislocation interaction energy. Our GNN model efficiently predicts these properties, reducing reliance on costly brute-force calculations and alleviating the computational demands on the multi-agent system. By combining the predictive capabilities of GNNs with the collaborative intelligence of LLM-driven reasoning agents, the system autonomously explores vast alloy design spaces, identifies trends in atomic-scale properties, and predicts macroscale mechanical strength, as demonstrated by several computational experiments. This synergistic approach accelerates the discovery of advanced alloys and holds promise for broader applications in other complex systems, marking a step forward in automated materials discovery and design. Impact statement Traditional deep learning models, such as graph neural networks and convolutional neural networks, operate within the confines of their training data sets, making single-step inferences for regression or classification. Our work introduces a multi-agent strategy that transcends these limitations by integrating deep learning with reasoning and decision-making capabilities. This intelligent system actively interprets results, determines subsequent actions, and iteratively refines predictions, accelerating the materials design process. We demonstrate its effectiveness in exploring the vast compositional space of a ternary alloy, where the model dynamically solicits data, analyzes trends, generates visualizations, and derives insights into materials behavior. By enabling accurate predictions of key alloy characteristics, our approach advances the discovery of novel metallic systems and underscores the critical role of solid-solution alloying. More broadly, it represents a major step toward integrating artificial intelligence with scientific reasoning, moving closer to artificial general intelligence in engineering. This paradigm shift has profound implications for materials science, enabling more efficient, autonomous, and intelligent exploration of complex materials spaces. Graphical Abstract

Clinical nutrition (CN) is becoming increasingly complex because of the rising prevalence of chronic illness, cancer, and malnutrition-related conditions such as short bowel syndrome and refeeding syndrome. Despite its clinical significance, nutrition education among US physicians remains limited. Simultaneously, large language model (LLM)-based artificial intelligence assistants (AIAs) have emerged as tools to support complex clinical decision-making but remain largely untested in CN. This retrospective study evaluated four LLM-based AIAs-ChatGPT (OpenAI), OpenEvidence (OpenEvidence Inc), Gemini (Google, Google DeepMind), and Copilot (Microsoft Corporation)-using five complex CN cases from our nutrition support service. Each AIA was queried with patient-specific CN questions. Responses were blinded and reviewed by five physician CN experts using an eight-item assessment tool evaluating clarity, relevance, evidence, and clinical utility. All AIAs produced clinically appropriate responses, with Gemini scoring highest in relevance (4.04) and clarity (4.16). Overall satisfaction scores ranged from 3.08 (Copilot) to 3.84 (Gemini). Citation quality and originality of insights varied and were generally limited, and no consistent differences in performance were observed across the five cases among the four AIAs. LLM-based AIAs can reliably replicate expert reasoning in CN. Although not yet a source of novel clinical insights, the true potential of this approach may lie in its application among physicians without specialized expertise in CN, helping to bridge existing knowledge gaps in nutrition care. Presenting full clinical cases, as shown in this study, could support AIA-enabled e-consultation in the future, thereby addressing gaps in CN education.

The proactive adoption of Generative Artificial Intelligence (GenAI) by e-commerce platforms to enhance consumer experience is emerging as a predominant trend. This research investigates the influence of AI overview on consumers’ perceived usefulness of the customer reviews section on e-commerce platforms, thereby further expanding the scope of application of the technology acceptance model (TAM). Across three scenario-based experiments (n = 568), we examined the effects of AI overview and their underlying mechanisms. Results consistently confirmed a main effect: the presence of AI overview significantly enhanced perceived usefulness compared to its absence. Study 2 identified perceived diagnosticity as a mediator, while Study 3 revealed that need for cognition (NFC) moderated both the main effect and the mediation process. Specifically, for High-NFC participants, the presence or absence of AI overview made no significant difference, whereas for Low-NFC participants, AI overviews significantly increased perceived usefulness. These findings offer novel insights into the effectiveness of AI overview in shaping the consumer evaluations of online customer reviews. By clarifying the mediating role of perceived diagnosticity and the boundary condition of NFC, this study contributes to a more nuanced understanding of how AI can be strategically integrated into e-commerce platforms to enhance consumer decision-making and guide business development.

With generative artificial intelligence (AI) completely changing the business landscape, it is no surprise that ethical leadership must also evolve in the corporate setting in response to this technological advancement. A framework for effectively navigating the uncertainty and complexity that AI presents is necessary to ensure that this innovation is used responsibly by the people at the top. This paper explores the ethical considerations between AI and ethical business leadership, specifically generative AI and its effects on corporate decision-making and corporate culture. Drawing on the ethical evaluation framework used with facial recognition technology (FRT), the research uses its principles to assess the ethical use of AI in corporate leadership. It examines how leaders strike a balance between innovation and responsibility in responding to AI-related challenges. This paper is grounded in servant leadership theory as a means of establishing a base of non-negotiables in how AI is used by leadership in corporate settings. The theory is grounded in seven principles: listening, empathy, healing, awareness, persuasion, conceptualization, and foresight. Servant leadership theory, compared to other ethical leadership models, such as transformational leadership and authentic leadership theory, prioritizes stakeholders' well-being, which is critical when integrating AI into corporate decision-making that directly impacts employees. The framework aims to create a tangible baseline for ethical leadership to use so that the implementation of generative AI in the workplace is not only impactful and receptive among employees but ultimately ethical. It enables leaders to interact with this technology in a way that prioritizes accountability, responsibility, and integrity in modern business environments.

This study investigates the adoption of generative artificial intelligence (AI) technologies—such as ChatGPT and DeepSeek—and their impact on employee well-being within university settings. Drawing on the Technology Acceptance Model (TAM), the research explores how perceived usefulness, ease of use, and enjoyment influence the adoption of AI tools by administrative staff. A quantitative survey was conducted with 164 university employees using a structured questionnaire. The findings revealed that perceived enjoyment is the only statistically significant predictor of AI adoption, while perceived usefulness and ease of use did not show significant effects. Furthermore, the adoption of generative AI tools was positively associated with employee happiness and negatively associated with stress levels. These results highlight the importance of intrinsic motivation and user experience in driving technology acceptance, especially in voluntary-use contexts. The study provides practical insights for university administrators seeking to enhance both AI adoption and employee well-being. The study recommends examining external factors such as organizational support, peer influence, and corporate culture to uncover how social and environmental conditions affect AI adoption. Equally important is addressing the potential downsides of AI, including user anxiety, job insecurity, and ethical concerns.

Artificial intelligence (AI) and big data are technologies with the potential to transform transfusion medicine (TM). This survey explored the scope of AI and big data use in TM across the Eastern Mediterranean and North Africa region. A survey was distributed among transfusion professionals to explore current use, perceived benefits and barriers to adopting AI and big data. Fifty respondents participated; the majority worked in national/regional transfusion services, and 58% worked in academic institutions. Only 24% reported using AI in daily TM practice, primarily for administrative tasks, education and research. Clinical applications were mainly in blood donor recruitment and management. Most used generative AI tools (92%) and were self-taught. Big data were employed in 36% of respondents' institutions, most often for inventory forecasting and optimizing blood product utilization. Most institutions used data based on laboratory information systems (89%), donor databases (72%) and electronic healthcare/patient records (67%). The main challenges and concerns regarding AI adoption were the lack of regulatory guidance, limited expertise, insufficient clinical validation of AI tools, implementation cost and ethical and privacy concerns. In terms of big data, the key barriers were insufficient expertise in data management and a lack of infrastructure for data storage. AI and big data adoption in TM within the region remains limited. Major barriers include regulatory gaps, lack of expertise, cost constraints and infrastructure limitations. Strategic investment in regulatory frameworks, targeted training and technical resources is essential to facilitate safe and effective integration into transfusion practice.

Artificial intelligence (AI) is increasingly involved in the film and television industry, and the relationship between process innovation, commercial value, and artistic value is worth exploring. As generative AI models and computer vision mature, the industry is transitioning from the "digital era" to the "intelligent era." This paper addresses the following questions: How does AI reconstruct the production chain and organizational structures of film and television? In addition to cost reduction and efficiency improvement, how does AI create new business models and market opportunities? How does AI participation in creative work reshape the core of artistic value and its evaluative standards? And are these dimensionsprocess, business, and artdriven in a linear fashion, or do they interact in complex and sometimes contradictory ways? Employing literature review, case studies, and expert interviews, this paper seeks to reveal the multidimensional impact of AI on the film and television industry and proposes implications for enterprise strategy and cultural governance.

Purpose The rise of artificial intelligence (AI) is transforming entire industries. Its impact on innovation management is receiving increasing attention. This study aims to answer the question of how the adoption of generative AI (GenAI) affects innovation management. Design/methodology/approach Our study identifies influencing factors from literature and links them to a multiple case study approach involving 13 organisations of different industries and sizes to gain more information on (1) how organisations use GenAI within their innovation management and (2) further internal and external influencing factors for a successful implementation. Findings Our empirical results reveal several motivating factors for companies to adopt GenAI, show distinct types of challenges (user-, technology-, company- and society-related) to cope with and unveil different ways in which companies implement GenAI in their innovation management. Based on the company's human focus and technological readiness regarding GenAI, a typology is derived that includes the four types of (1) explorers, (2) technology specialists, (3) human-centred visionaries and (4) combiners. The overall cross-case comparison outlines further aspects of the successful implementation of GenAI. Originality/value Our study outlines similarities and differences of GenAI adoption in innovation management and provides valuable new insights. The study thus offers guidance for researchers, organisations and policymakers and provides implications for future research.