BiXiao: An AI-dirven Atmospheric Environmental Forecasting Model with Non-continuous Grids

Although artificial intelligence (AI) is currently a prominent technology that has received increasing attention, the factors affecting users’ AI device adoption for healthcare were less discussed. Drawing on affordance-actualization theory and literature on AI, uniqueness theory, and healthcare, we developed a research model to explain users’ adoption intention of AI devices in the healthcare context. Data collected from 307 potential users in China were used to test the proposed model. Our findings indicated that data capture, classification, and social affordances impacted self-monitoring significantly, while data capture and social affordances were the predators of uniqueness enhancement. The self-monitoring, on the other hand, had a positive influence on uniqueness enhancement. Finally, our results revealed that self-monitoring and uniqueness enhancement have significant effects on AI adoption. From the findings, theoretical and practical implications are discussed.

Although Artificial Intelligence (AI) is being increasingly applied, considerable distrust about introducing “disruptive” technologies persists. Intrinsic and contextual factors influencing where and how such innovations are introduced therefore require careful scrutiny to ensure that health equity is promoted. To illustrate one such critical approach, we describe and appraise an AI application – the development of computer assisted diagnosis (CAD) to support more efficient adjudication of compensation claims from former gold miners with occupational lung disease in Southern Africa. In doing so, we apply a bio-ethical lens that considers the principles of beneficence, non-maleficence, autonomy and justice and add explicability as a core principle. We draw on the AI literature, our research on CAD validation and process efficiency, as well as apprehensions of users and stakeholders. Issues of concern included AI accuracy, biased training of AI systems, data privacy, impact on human skill development, transparency and accountability in AI use, as well as intellectual property ownership. We discuss ways in which each of these potential obstacles to successful use of CAD could be mitigated. We conclude that efforts to overcoming technical challenges in applying AI must be accompanied from the onset by attention to ensuring its ethical use.

Although artificial intelligence (AI) has quickly become a tool that offers quick, simplistic answers to reference questions (Chen, 2023), in this paper, we interrogate AI’s ability to consider the human-to-human interpersonal nuances that occur and are vital to answering queries. We ask: what is most essential for reference services - human knowledge sharing or the machine’s artificial intelligence? AI, as a mechanistic language model, cannot discern the subtleties of humanity’s questions; AI itself admits that librarians are more efficient at answering reference queries from a humanistic approach and lens (Yang & Mason, 2023). This paper is the second part of our introduction to our methodology for reference services, the Query Search Method (QSM). We present conceptual substantiation that, when enacted, the QSM can incorporate AI as an addition to the librarian’s toolbox for honoring patrons’ ways of knowing that, through the sententiousness of the human experience, invariably seek knowledge beyond “the machine.”

AI disruption threat and employee outcomes: Role of technology insecurity, thriving at work, and trait self-esteem

Effects of chemical and meteorological data assimilation on air-quality and meteorological forecasts in the Korean Peninsula.

The 21st century has ushered in incredibly intelligent machines and software capable of doing almost anything. While some might argue that this makes young people less inclined to develop certain skills, there is no doubt that these machines can handle complex scientific thinking and tedious calculations with just a single click. Although artificial intelligence (AI) has been around for a while, recent breakthroughs and increased availability have made it useful in all areas of science. This technology is revitalizing and revolutionizing the world. In a few years, greater equalization will change the dynamics of our present reality. AI has a major impact on industries by creating new ideas, making things more productive, and changing the way businesses and society function. Additionally, AI is helping scientific research by accelerating discoveries, making it possible to run intricate simulations and experiments, and providing people with new abilities.

Although artificial intelligence (AI), the leading technology in this digital era, has become an important driver for business investments, little is known about its underlying theoretical mechanisms and functioning boundaries. Drawing on the institution-based view, we argue that AI development creates an emerging institution, serving as a centripetal force to attract foreign firm investment (FDI). Accordingly, we predict that host-country AI development attracts a firm’s FDI in the form of subsidiary establishment. We highlight home-subnational AI development and a firm’s state ownership as important boundary conditions under which the relationship is stronger and weaker, respectively. We construct a novel dataset reflecting rich AI development information, drawing upon a sample of 1,524 Chinese public firms engaged in outward FDI activities in 26 host countries from 2015 to 2017. We find evidence for our predictions. Our results provide valuable implications for the literature on international digital economy and AI research.

Purpose Big data analytics allows researchers and industry practitioners to extract hidden patterns or discover new information and knowledge from big data. Although artificial intelligence (AI) is one of the emerging big data analytics techniques, hospitality and tourism literature has shown minimal efforts to process and analyze big hospitality data through AI. Thus, this study aims to develop and compare prediction models for review helpfulness using machine learning (ML) algorithms to analyze big restaurant data. Design/methodology/approach The study analyzed 1,483,858 restaurant reviews collected from Yelp.com. After a thorough literature review, the study identified and added to the prediction model 4 attributes containing 11 key determinants of review helpfulness. Four ML algorithms, namely, multivariate linear regression, random forest, support vector machine regression and extreme gradient boosting (XGBoost), were used to find a better prediction model for customer decision-making. Findings By comparing the performance metrics, the current study found that XGBoost was the best model to predict review helpfulness among selected popular ML algorithms. Results revealed that attributes regarding a reviewer’s credibility were fundamental factors determining a review’s helpfulness. Review helpfulness even valued credibility over ratings or linguistic contents such as sentiment and subjectivity. Practical implications The current study helps restaurant operators to attract customers by predicting review helpfulness through ML-based predictive modeling and presenting potential helpful reviews based on critical attributes including review, reviewer, restaurant and linguistic content. Using AI, online review platforms and restaurant websites can enhance customers’ attitude and purchase decision-making by reducing information overload and search cost and highlighting the most crucial review helpfulness features and user-friendly automated search results. Originality/value To the best of the authors’ knowledge, the current study is the first to develop a prediction model of review helpfulness and reveal essential factors for helpful reviews. Furthermore, the study presents a state-of-the-art ML model that surpasses the conventional models’ prediction accuracy. The findings will improve practitioners’ marketing strategies by focusing on factors that influence customers’ decision-making.

Although artificial intelligence (AI) has had a profound impact on areas such as image recognition, comparable advances in drug discovery are rare. This article quantifies the stages of drug discovery in which improvements in the time taken, success rate or affordability will have the most profound overall impact on bringing new drugs to market. Changes in clinical success rates will have the most profound impact on improving success in drug discovery; in other words, the quality of decisions regarding which compound to take forward (and how to conduct clinical trials) are more important than speed or cost. Although current advances in AI focus on how to make a given compound, the question of which compound to make, using clinical efficacy and safety-related end points, has received significantly less attention. As a consequence, current proxy measures and available data cannot fully utilize the potential of AI in drug discovery, in particular when it comes to drug efficacy and safety in vivo. Thus, addressing the questions of which data to generate and which end points to model will be key to improving clinically relevant decision-making in the future.

<p dir="ltr"><span>Gestational Diabetes Mellitus (GDM) is characterized as any degree of glucose intolerance that manifests during pregnancy and typically resolves postpartum. Diagnosis is commonly made through an Oral Glucose Tolerance Test (OGTT); however, there are notable inconsistencies in diagnostic criteria and treatment thresholds both nationally and internationally. The onset of GDM generally occurs in the late second or early third trimester, with potential complications including fetal macrosomia, which may lead to difficult labor, and neonatal hypoglycemia due to excess insulin production in the infant’s pancreas. Furthermore, certain studies suggest that GDM may delay fetal brain development, resulting in long-term neurological impairments. Although artificial intelligence (AI) models have only recently emerged as a potential solution in various medical fields, their application continues to raise concerns, particularly regarding the use and storage of personal data. Machine Learning (ML), a subset of AI, utilizes multivariate classification methods, also known as supervised pattern recognition approaches, which are designed to identify patterns and correlations among multiple variables to categorize them into specific groups or classes. In this context, we are describing here .Thus, artificial intelligence has not yet demonstrated its full clinical potential in the management of gestational diabetes mellitus (GDM). Current applications remain limited in their impact on improving patient outcomes, largely due to methodological heterogeneity and the early stage of implementation. Still, AI has shown considerable promise in the domain of predictive modeling, particularly through its ability to process large volumes of clinical and biochemical data. This analytical capacity enables the identification of complex patterns and correlations that can support the accurate early prediction of GDM, potentially allowing for earlier intervention and personalized care strategies. Future research should focus on validating these predictive models in larger, diverse populations and integrating them into clinical workflows under appropriate regulatory and ethical frameworks.</span></p>

Objectives Although Artificial Intelligence (AI) has been a part of the computer science field for many decades, it has only recently been applied to different areas of behavioral and social sciences. This article provides an examination of the applications of AI methodologies to behavioral and social sciences exploring the areas where they are now utilized, the different tools used and their effectiveness. Methods The study is a systematic research examination of peer-reviewed articles (2010–2019) that used AI methodologies in social and behavioral sciences with a focus on children and families. Results The results indicate that artificial intelligence methodologies have been successfully applied to three main areas of behavioral and social sciences, namely (1) to increase the effectiveness of diagnosis and prediction of different conditions, (2) to increase understanding of human development and functioning, and (3) to increase the effectiveness of data management in different social and human services. Random forests, neural networks, and elastic net are among the most frequent AI methods used for prediction, supplementing traditional approaches, while natural language processing and robotics continue to increase their role in understanding human functioning and improve social services. Conclusions Applications of AI methodologies to behavioral and social sciences provide opportunities and challenges that need to be considered. Recommendations for future research are also included.KeywordsArtificial intelligenceBehavioral and social sciencesMachine learningFamiliesChildren

How Do Clinicians Perceive the Role of AI in Cardiovascular Disease Prevention for Women?

Purpose Although artificial intelligence (AI) streamers offer promising solutions to the limitation of human streamers, effectively designing them to create emotional connections with consumers remains challenging. This study addresses a literature gap by extending Kindchenschema theory, which has primarily been studied in static product design, to the dynamic and multi-modal domain of AI streamer design in livestreaming commerce. Design/methodology/approach This research investigates how cute AI streamers influence consumer behavior whilst livestreaming commerce using three experimental studies. Data were collected through online experiments with diverse consumer samples and analyzed using the PROCESS macro for SPSS to test the mediating and moderating effects with bootstrap confidence intervals. Findings Study 1 demonstrates that the cuteness of AI streamers enhances purchase intentions. Study 2 tests the main effects of cuteness on purchase intentions through the parallel mediating effects of delight and perceived friendliness. Study 3 examines the moderating role of the autotelic need for touch, revealing that the effect of cuteness on purchase intentions is stronger for consumers with a high need for touch. Originality/value This research extends Kindchenschema theory to dynamic AI interactions and develops a multi-modal framework for cuteness design. The findings provide actionable guidance for companies creating interactive AI streamers in the realm of digital commerce, suggesting adaptable cuteness strategies that can be tailored according to service context and product categories to enhance purchasing decisions in digital marketing settings.

Mental health providers (MHPs) face a significant administrative burden from documentation, which can contribute to burnout and reduce time available for direct patient care. Although artificial intelligence (AI)-powered scribes have shown promise in general medical settings, their utility has not been well explored in the specific context of mental health care. This study describes the development and preliminary observational evaluation of Smart Notes, a generative AI tool designed to assist MHPs with documentation on a commercial virtual mental health platform. This study aimed to examine MHPs' use and uptake of a generative AI documentation tool, including use patterns, perceptions of note quality, feedback and satisfaction, and changes in productivity. Smart Notes was developed using a HIPAA (Health Insurance Portability and Accountability Act)-compliant Azure OpenAI infrastructure to securely generate session summaries from individual therapy sessions. The tool was rolled out to MHPs in a phased approach, and its use required MHP and client consent as well as a mandatory review and edit by the MHP. We conducted a 1-year retrospective observational evaluation of the feature, examining MHP use, MHP-rated note quality, MHP feedback and satisfaction, and MHP productivity (weekly working hours, session completion rate, and client caseload). Over 1 year, 162 full-time and 1366 contractual MHPs used Smart Notes to generate over 286,000 clinical notes. Use of the feature was high and stable, with nearly all (averaging 94% each week, SD 1%) full-time MHPs and most (averaging 72% each week, SD 6%) contractual MHPs using it weekly for eligible individual therapy sessions after the full launch. MHP-rated note quality was overwhelmingly positive, with 97.7% (9980/10,219) of feedback ratings among full-time MHPs and 98.4% (19,977/20,300) of feedback ratings among contractual MHPs being a "thumbs-up." Qualitative feedback from MHPs was also largely positive, praising the tool for saving time and easing administrative burden. Finally, both full-time and contractual MHPs demonstrated changes in productivity. Our findings suggest that AI-powered documentation tools, such as Smart Notes, are a feasible and acceptable approach to supporting MHPs. The high adoption rate and favorable MHP feedback indicate the potential utility of these tools without compromising note quality. This study provides preliminary data regarding the application of AI documentation in the mental health care context, highlighting a promising path for future research and development in digital mental health.

The co‑epidemic of silicosis and tuberculosis (TB) in South Africa’s mining industry affects a large number of migrant workers and is compounded by limited access to chest X‑ray (CXR) screening. Although artificial intelligence (AI)‑based computer‑aided detection (CAD) systems for TB have demonstrated impressive accuracy against microbiological standards, validation among silica‑exposed populations has been limited. Moreover, well‑documented biases hinder CAD utility in diverse patient populations, potentially exacerbating existing healthcare inequities. In this article, we describe the challenges in developing CAD systems for TB and silicosis and present the potential benefits local public‑sector development initiatives can bring.Using a local dataset of 2000 CXRs from silica‑exposed Southern African mineworkers, alongside publicly available international datasets and pretrained CAD models, we present empirical evidence of CAD biases. Dimensionality reduction analysis produced visual mappings that demonstrate how local CXRs form a distinct cluster, separate from international images. We also found that, relative to TB, reducing image resolution disproportionately degraded silicosis detection. Further visualizations proved that accuracy metrics alone are insufficient measures of clinical reliability, possibly obscuring deployment failures.We conclude that local public‑sector CAD development offers a viable alternative to reliance on externally developed systems that likely exclude underserved populations. Addressing CAD deficiencies requires curating population‑representative datasets that capture local epidemiology and transparent, open‑source development practices that enable peer review and bias correction. Embedding technical and clinical expertise locally can transform AI‑based CAD from a potential instrument of digital colonialism into a mechanism that produces contextually appropriate diagnostics while advancing knowledge for equitable AI deployment worldwide.