Artificial Intelligence Products Research Articles

AI in medical education: the moderating role of the chilling effect and STARA awareness

BackgroundThe rapid growth of artificial intelligence (AI) technologies has been driven by the latest advances in computing power. Although, there exists a dearth of research on the application of AI in medical education.Methodsthis study is based on the TAM-ISSM-UTAUT model and introduces STARA awareness and chilling effect as moderating variables. A total of 657 valid questionnaires were collected from students of a medical university in Dalian, China, and data were statistically described using SPSS version 26, Amos 3.0 software was used to validate the research model, as well as moderated effects analysis using Process (3.3.1) software, and Origin (2021) software.ResultsThe findings reveal that both information quality and perceived usefulness are pivotal factors that positively influence the willingness to use AI products. It also uncovers the moderating influence of the chilling effect and STARA awareness.ConclusionsThis suggests that enhancing information quality can be a key strategy to encourage the widespread use of AI products. Furthermore, this investigation offers valuable insights into the intersection of medical education and AI use from the standpoint of medical students. This research may prove to be pertinent in shaping the promotion of Medical Education Intelligence in the future.

The Synergy of Ambidextrous Leadership, Agility, and Entrepreneurial Orientation to Achieve Sustainable AI Product Innovation

This study aims to explore potential mechanisms of ambidextrous leadership (AL) in product innovativeness from the perspective of organizational agility (OA) and entrepreneurial orientation (EO) in firms operating in the artificial intelligence (AI) industry. A quantitative research method was used with 405 questionnaires, and the respondents were randomly selected from reputable databases. Structural equation modeling was employed to evaluate the model fit and conduct hypothesis testing. The findings suggest that ambidextrous leadership demonstrates a significant positive influence on product innovativeness and OA; also, through the mediating role of OA, it is possible to analyze both the direct and indirect relationships among the factors. Additionally, the moderating effect of EO on the intercorrelations among these factors was explored. This study enhances existing knowledge on leadership dynamics in the context of new product development, highlights the importance of adaptability in leadership, and sheds light on the interplay between OA, EO, and new product innovation. This study highlights the role of product innovativeness in sustainable AI product development. Enhanced product innovativeness not only sustains AI product development but also promotes environmental sustainability. This is achieved through the minimization of energy use, reduction in material requirements, and prevention of pollution. Firms are using these insights to develop sustainable and eco-friendly products, as well as create new market opportunities while reducing environmental impact. This research underscores the interconnectedness of factors in this study and sustainability, providing a new perspective on sustainable AI product development.

A growing number of AIs cleared for clinical use is finally available: The AI-assisted Pathologist

As more pathology laboratories are transitioning to a digital workflow, the availability of commercial Artificial Intelligence assistance systems is also increasing. Today nearly 40 such products approved for diagnostic use are available. This article provides an overview of the most widely addressed use cases, including Immunohistochemistry scoring for breast cancer and non-small-cell lung cancer, Gleason grading for prostate cancer, or metastasis detection in lymph nodes. While automation alone already promises an increase in efficiency that may help to bridge the growing gap between supply (pathology work force) and demand (histological testing), this article introduces another category of Artificial Intelligence products that go beyond just mimicking today’s established score. Various Artificial Intelligence are being introduced that detect genetic alterations or stratify risk, directly from the standard hematoxylin & eosin staining. Finally, a brief outlook explains how basic AI models are currently finding their way into computational pathology and promise to further accelerate product developments by decreasing the time-to-model.

On the Copyright Analysis of Artificial Intelligence Products

With the gradual maturity of artificial intelligence technology, artificial intelligence products continue to emerge in the market, which poses new challenges to the current copyright law system. In particular, whether artificial intelligence products are copyrightable is a hot and difficult issue in the current theoretical and practical circles. Based on this, there are two mainstream viewpoints in the current theoretical circle, that is, the viewpoint that supports copyright protection for artificial intelligence products from the perspective of "reader-centered", and the viewpoint that opposes copyright protection for artificial intelligence products from the perspective of author-centered". However, the question of whether artificial intelligence products can be protected by copyright involves verifying the reasonableness and forward-looking nature of the "open" regulation of copyright objects in China on one hand. On the other hand, it is also relates to determining the future legislative direction of high-level artificial intelligence. Therefore, by comparing the similarities and differences in originality between the copyright law system and the copyright law system, the conclusion is drawn that artificial intelligence products should be protected by copyright. The evaluation criteria of subject-object separation established by "reader-centered doctrine" should be adopted.

Dual Sensing Signal Decoupling Based on Thermoelectric Polymer Aerogels for Precise Temperature and Pressure Recognition

AbstractThe capability to emulate skin‐like temperature and pressure sensing is fundamental for next‐generation artificial intelligence products. However, detecting temperature and pressure simultaneously with a single sensor without signal interference is challenging. Herein, a novel PCC aerogel sensor composed of PEDOT:PSS, CNTs, and CNF via directional freezing technology is developed. The PCC sensor can decouple temperature and pressure stimuli into individual voltage and resistance signals. It exhibits high‐precision temperature sensing capabilities, boasting an exceptionally high Seebeck coefficient of 30.4 µV K‐1 and the ability to detect temperature variations as low as 0.1 K. PCC sensors show excellent sensitivity and fast response times for detecting static and dynamic pressures, as well as high stability after 1000 cycles. Its maximum pressure sensitivity can reach 159.1% kPa−1, and the lowest detection limit is 10 Pa. Additionally, its excellent thermoelectric properties also enable to generating thermopower from human skin for self‐powered pressure sensing. A 3×3 PCC sensor array has been proposed to simulate the unique features of human skin in temperature and pressure recognition. This work provides a scalable manufacturing strategy for multi‐functional tactile sensors.

Co‐designing enduring learning analytics prediction and support tools in undergraduate biology courses

AbstractEven highly motivated undergraduates drift off their STEM career pathways. In large introductory STEM classes, instructors struggle to identify and support these students. To address these issues, we developed co‐redesign methods in partnership with disciplinary experts to create high‐structure STEM courses that better support students and produce informative digital event data. To those data, we applied theory‐ and context‐relevant labels to reflect active and self‐regulated learning processes involving LMS‐hosted course materials, formative assessments, and help‐seeking tools. We illustrate the predictive benefits of this process across two cycles of model creation and reapplication. In cycle 1, we used theory‐relevant features from 3 weeks of data to inform a prediction model that accurately identified struggling students and sustained its accuracy when reapplied in future semesters. In cycle 2, we refit a model with temporally contextualized features that achieved superior accuracy using data from just two class meetings. This modelling approach can produce durable learning analytics solutions that afford scaled and sustained prediction and intervention opportunities that involve explainable artificial intelligence products. Those same products that inform prediction can also guide intervention approaches and inform future instructional design and delivery.Practitioner notesWhat is already known about this topic Learning analytics includes an evolving collection of methods for tracing and understanding student learning through their engagements with learning technologies. Prediction models based on demographic data can perpetuate systemic biases. Prediction models based on behavioural event data can produce accurate predictions of academic success, and validation efforts can enrich those data to reflect students' self‐regulated learning processes within learning tasks. What this paper adds Learning analytics can be successfully applied to predict performance in an authentic postsecondary STEM context, and the use of context and theory as guides for feature engineering can ensure sustained predictive accuracy upon reapplication. The consistent types of learning resources and cyclical nature of their provisioning from lesson to lesson are hallmarks of high‐structure active learning designs that are known to benefit learners. These designs also provide opportunities for observing and modelling contextually grounded, theory‐aligned and temporally positioned learning events that informed prediction models that accurately classified students upon initial and later reapplications in subsequent semesters. Co‐design relationships where researchers and instructors work together toward pedagogical implementation and course instrumentation are essential to developing unique insights for feature engineering and producing explainable artificial intelligence approaches to predictive modelling. Implications for practice and/or policy High‐structure course designs can scaffold student engagement with course materials to make learning more effective and products of feature engineering more explainable. Learning analytics initiatives can avoid perpetuation of systemic biases when methods prioritize theory‐informed behavioural data that reflect learning processes, sensitivity to instructional context and development of explainable predictors of success rather than relying on students' demographic characteristics as predictors. Prioritizing behaviours as predictors improves explainability in ways that can inform the redesign of courses and design of learning supports, which further informs the refinement of learning theories and their applications.

Protocol for evaluating the fitness for purpose of an artificial intelligence product for radiology reporting in the BreastScreen New South Wales breast cancer screening programme

IntroductionRadiologist shortages threaten the sustainability of breast cancer screening programmes. Artificial intelligence (AI) products that can interpret mammograms could mitigate this risk. While previous studies have suggested this technology has...

Big Data Analysis and Deep Learning Optimization in Artificial Intelligence Production of Information Enterprises

Intelligent manufacturing technology is required to upgrade existing enterprises' management and production operations. To construct a ground breaking fusion structure, this project unites the theoretical underpinnings, technical breakthroughs, and applications of data analytics, optimisation, and intelligent production engineering. It is driven by China's desire of cutting-edge commodities and efficient growth methods. This research establishes the broad framework for merging optimisation and data analytics. There is a list of data analytics and system optimisation technologies that can address important challenges with intelligent manufacturing. By integrating data analytics and optimisation, businesses may better forecasting and management of new terrain, as well as reveal hidden information to increase decision-making efficacy.

The advent of ChatGPT: Job Made Easy or Job Loss to Data Analysts

Artificial Intelligence (AI) has proven valuable in almost every field of endeavour, including education, sciences, engineering, technology, medical sciences, and numerous other areas of application. Despite its widespread usefulness, concerns have arisen about AI potentially displacing jobs due to its highly advanced capabilities, commonly called the "god-man effect." One remarkable AI product is ChatGPT, a Chatbot developed by an open AI company in the USA that is capable of engaging in conversations that resemble human interactions. This study explores the strengths and limitations of ChatGPT for data analysis, with the primary objective of assessing whether ChatGPT poses a threat to the job of data analysts. An econometric dataset with a sample size of thirty (30), which consists of one dependent variable and three independent variables, was simulated. The dataset was intentionally generated with issues like multicollinearity, outliers, and heteroscedasticity. Subsequently, multiple tests were conducted on the datasets to confirm the presence of these problems. The ChatGPT 3.5 and 4.0 versions were then used to analyse the data to examine this chatbot's prowess in performing data analysis. ChatGPT 3.5 and 4.0 accurately predicted the suitable statistical tool for analyzing the simulated datasets. Both versions of ChatGPT emphasized that the expertise of a professional data analyst would be necessary. While they could offer guidance on data analysis, they cannot perform the analysis themselves as they are solely AI models. ChatGPT can help with what to do next when a data analyst gets stuck. However, they should not be recognized as an authority in making statistical decisions. Therefore, ChatGPT may not replace data analysts but could make their job easier by serving as a helpful resource to turn to when they encounter challenges.

Human bias in evaluating AI product creativity

This study investigated the evaluation behavior of products generated by artificial intelligence (AI) by manipulating the creativity of rating stimuli according to their high and low novelty and usefulness. In addition, we analyzed the effect of perception of the generative AI on creativity ratings. We found a bias toward lower creativity ratings only for highly useful products when they were identified as AI-generated, and this bias was particularly strong for individuals with greater perceived threat from the generative AI. Our research highlights the importance of product attribute quality and producer identity in creativity ratings, and provides implications for the integration of AI into creative industries.

Considering the Secondary Use of Clinical and Educational Data to Facilitate the Development of Artificial Intelligence Models.

Medical training programs and health care systems collect ever-increasing amounts of educational and clinical data. These data are collected with the primary purpose of supporting either trainee learning or patient care. Well-established principles guide the secondary use of these data for program evaluation and quality improvement initiatives. More recently, however, these clinical and educational data are also increasingly being used to train artificial intelligence (AI) models. The implications of this relatively unique secondary use of data have not been well explored. These models can support the development of sophisticated AI products that can be commercialized. While these products have the potential to support and improve the educational system, there are challenges related to validity, patient and learner consent, and biased or discriminatory outputs. The authors consider the implications of developing AI models and products using educational and clinical data from learners, discuss the uses of these products within medical education, and outline considerations that should guide the appropriate use of data for this purpose. These issues are further explored by examining how they have been navigated in an educational collaborative.

Applications of Deep Learning in Trauma Radiology: A Narrative Review

Diagnostic imaging is essential in modern trauma care for initial evaluation and identifying injuries requiring intervention. Deep learning (DL) has become mainstream in medical image analysis and has shown promising efficacy for classification, segmentation, and lesion detection. This narrative review provides the fundamental concepts for developing DL algorithms in trauma imaging and presents an overview of current progress in each modality. DL has been applied to detect free fluid on Focused Assessment with Sonography for Trauma (FAST), traumatic findings on chest and pelvic X-rays, and computed tomography (CT) scans, identify intracranial hemorrhage on head CT, detect vertebral fractures, and identify injuries to organs like the spleen, liver, and lungs on abdominal and chest CT. Future directions involve expanding dataset size and diversity through federated learning, enhancing model explainability and transparency to build clinician trust, and integrating multimodal data to provide more meaningful insights into traumatic injuries. Though some commercial artificial intelligence products are Food and Drug Administration-approved for clinical use in the trauma field, adoption remains limited, highlighting the need for multi-disciplinary teams to engineer practical, real-world solutions. Overall, DL shows immense potential to improve the efficiency and accuracy of trauma imaging, but thoughtful development and validation are critical to ensure these technologies positively impact patient care.

Product information recognition in the retail domain as an MRC problem

This paper presents the task of recognizing product information (PI) (i.e., product names, prices, materials, etc.) mentioned in customer statements. This is one of the key components in developing artificial intelligence products to enable businesses to listen to their customers, adapt to market dynamics, continuously improve their products and services, and improve customer engagement by enhancing effectiveness of a chatbot. To this end, natural language processing (NLP) tools are commonly used to formulate the task as a traditional sequence labeling problem. However, in this paper, we bring the power of machine reading comprehension (MRC) tasks to propose another, alternative approach. In this setting, determining product information types is the same as asking “Which PI types are referenced in the statement?” For example, extracting product names (which corresponds to the label PRO_NAME) is cast as retrieving answer spans to the question “Which instances of product names are mentioned here?” We perform extensive experiments on a Vietnamese public dataset. The experimental results show the robustness of the proposed alternative method. It boosts the performance of the recognition model over the two robust baselines, giving a significant improvement. We achieved 92.87% in the F1 score on recognizing product descriptions at Level 1. At Level 2, the model yielded 93.34% in the F1 score on recognizing each product information type.

A comprehensive approach to safety for highly automated off-road machinery under Regulation 2023/1230

As of January 2027, Machine Safety Regulation 2023/1230 replaces Directive 2006/42/EC, and a proposal for the regulation of Artificial Intelligence (AI) systems is soon to be ratified by the European Council. These legislative changes address hazards from increased automation and AI integration. Original Equipment Manufacturers (OEMs) face compliance challenges due to a lack of standardized guidelines, especially for off-road machinery with level 4 ‘high automation’ capabilities.At level 4 ‘high automation’, the machine operator assumes a supervisory role, intervening only in situations beyond the designed operational domain. Relying solely on supervisor intervention at this level of automation is unreasonable. Within this context, legislative safety requirements to maintain safety advocate for robust risk management systems, including both a human-in-the-loop safety option and adequate safety-related decision-making by the level 4 ‘high automation’ off-road machine, i.e., the safe reasoning software becomes paramount.In this research, conducted through a systematic mapping study, we discuss emerging safety trends and integration for the safe reasoning software a level 4 off-road machine requires. Our analyses shows the necessity of safety constraint-based design procedures. These procedures involve constraining the AI system by embedding fault management, reasoning checks, and diagnostic monitoring together with safety-related decision-making, thereby enhancing functional reliability and reducing risks. We consider these trends and patterns through the lens of changing machine safety legislative requirements, and the proposal of the regulation of AI products. Illustrating their implications alongside a case study of a level 4 tree harvesting machine, for a tangible and to industry transferable perspective on our results.

Feedback Beamforming in the Time Domain.

Real-time source localization is crucial for high-end automation and artificial intelligence (AI) products. However, a low signal-to-noise ratio (SNR) and limited processing time can reduce localization accuracy. This work proposes a new architecture for a time-domain feedback-based beamformer that meets real-time processing demands. The main objective of this design is to locate reflective sources by estimating their direction of arrival (DOA) and signal range. Incorporating a feedback mechanism in this architecture refines localization precision, a unique aspect of this approach. We conducted an in-depth analysis to compare the effectiveness of time-domain feedback beamforming against conventional time-domain methods, highlighting their benefits and limitations. Our evaluation of the proposed architecture, based on critical performance indicators such as peak-to-sidelobe ratio, mainlobe width, and directivity factor, demonstrates its ability to improve beamformer effectiveness significantly.

Research on Machine Learning Copyright

A new round of scientific and technological revolution and industrial transformation led by artificial intelligence is on the rise. Driven by new theories and technologies, artificial intelligence presents new features such as deep learning, cross-border integration, human-machine collaboration, swarm intelligence openness, and autonomous control, and is having a significant and far-reaching impact on economic development, social progress, and global governance. Artificial intelligence technology has made a leap forward, and the ensuing legal issues have aroused discussions in the academic and practical circles, mainly focusing on the copyright law of artificial intelligence products, including the legal attribute of the generated results and its protection path, as well as exploring the new path of machine learning infringement risk and its compliance governance.

A multi-skilled staff scheduling and team configuration optimisation model for artificial intelligence project portfolio considering competence development and innovation-driven

This paper emphasises the pivotal role of enhancing research and development (R&D) staff competence and boosting team innovation efficiency in configuring R&D project teams for artificial intelligence (AI) product development. Diverging from traditional studies primarily focused on time, quality, and cost objectives, this study proposes prioritising the skill increments of staff and team diversity, aligning with the overarching goals of the R&D cycle. Targeting multi-skilled R&D personnel for scheduling and configuration, a three-objective tradeoff optimisation model is established. The nonlinear mixed-integer constrained programming model incorporates a learning effect for calculating employee skill value and employs a heterogeneity efficiency formula for assessing team diversity, particularly emphasising the diversity of R&D personnel research backgrounds. An algorithm based on Non-dominated Sorting Genetic Algorithm-II (NSGA-II) is designed to obtain the approximate Pareto optimal solution. Furthermore, it compares the algorithm with the multi-objective particle swarm optimisation (MOPSO), explores practical decision-making methods for Pareto solutions, and conducts sensitivity analyses on the learning rate and diversity level. Our research is relevant to enterprises seeking to enhance R&D capabilities with a certain degree of homogeneity among R&D employees. This paper exemplifies and validates the proposed model and solution approach using a new AI product from a healthcare company.

The impact of artificial intelligence technology innovation on economic development -- from the perspective of generative AI products

The new generation of artificial intelligence products, represented by ChatGPT and Wenyan Yixin, is rapidly and extensively integrating into human production and life at an unprecedented pace, breadth, and depth. Generative artificial intelligence possesses distinct characteristics and functionalities compared to rule-based artificial intelligence, exerting a more prominent dual impact on the high-quality development of China's economy. On the one hand, it can promote high-quality economic development through empowering effects such as innovation-driven effects, enhanced production efficiency, and industrial transformation and upgrading effects. On the other hand, it may give rise to deep-seated risks, including labor market disruptions, market monopolization issues, national security risks, and misinformation phenomena, which could hinder high-quality economic development. It is crucial to have a correct understanding and a scientific approach, leveraging its positive effects while proactively addressing negative impacts, guiding it to better serve the needs of China's high-quality economic development.

A typology of artificial intelligence data work

This article provides a new typology for understanding human labour integrated into the production of artificial intelligence systems through data preparation and model evaluation. We call these forms of labour ‘AI data work’ and show how they are an important and necessary element of the artificial intelligence production process. We draw on fieldwork with an artificial intelligence data business process outsourcing centre specialising in computer vision data, alongside a decade of fieldwork with microwork platforms, business process outsourcing, and artificial intelligence companies to help dispel confusion around the multiple concepts and frames that encompass artificial intelligence data work including ‘ghost work’, ‘microwork’, ‘crowdwork’ and ‘cloudwork’. We argue that these different frames of reference obscure important differences between how this labour is organised in different contexts. The article provides a conceptual division between the different types of artificial intelligence data work institutions and the different stages of what we call the artificial intelligence data pipeline. This article thus contributes to our understanding of how the practices of workers become a valuable commodity integrated into global artificial intelligence production networks.

Information communication technology, economic growth, natural resources, and renewable energy production: Evaluating the asymmetric and symmetric impacts of artificial intelligence in robotics and innovative economies

The ongoing integration of artificial intelligence (AI) into the fundamental fabric of society constitutes a transformative process, with the potential to exert profound influence on various facets of human existence including environmental quality. Our research investigates the interaction of AI and renewable energy production in 22 leading robotics and innovative countries, providing crucial insights into a previously underexplored domain. The study's novelty and originality are in demonstrating the precise impact of AI in enhancing and revolutionizing renewable energy production, consequently addressing, and filling a critical study gap. Our analysis comprises both asymmetric and symmetric assumptions to cover the positive and negative shocks of artificial intelligence on renewable energy from 1991 to 2020. The findings from the panel NARDL analysis provide substantial support for the profound impact of changes in artificial intelligence (AI) on the substantial stimulation of renewable energy in the long-run. Furthermore, AI plays a substantial role in enhancing long-term renewable energy production in the panel symmetric ARDL. In the country-specific investigation, it is found that Austria, Germany, and New Zealand are the nations where both affirmative and adverse changes in AI improve clean energy structure significantly in the short run. Natural resources, information communication technology (ICT), and economic growth also enhance renewable energy production. Finally, policies are directed to increase renewable energy production.