Multiscale Coordination Dynamics: Reflections and Opportunities for Advancing Groups and Teams Research

Multiscale systems that are characterized by a great range of spatial–temporal scales arise widely in many scientific domains. These range from the study of protein conformational dynamics to multiphase processes in, for example, granular media or haemodynamics, and from nuclear reactor physics to astrophysics. Despite the diversity in subject areas and terminology, there are many common challenges in multiscale modelling, including validation and design of tools for programming and executing multiscale simulations. This Theme Issue seeks to establish common frameworks for theoretical modelling, computing and validation, and to help practical applications to benefit from the modelling results. This Theme Issue has been inspired by discussions held during two recent workshops in 2013: ‘Multiscale modelling and simulation’ at the Lorentz Center, Leiden (http://www.lorentzcenter.nl/lc/web/2013/569/info.php3?wsid=569&venue=Snellius), and ‘Multiscale systems: linking quantum chemistry, molecular dynamics and microfluidic hydrodynamics’ at the Royal Society Kavli Centre. The objective of both meetings was to identify common approaches for dealing with multiscale problems across different applications in fluid and soft matter systems. This was achieved by bringing together experts from several diverse communities.

Relevance. The development of artificial intelligence (AI) as a societal foundation has become crucial for leading economies, who view it as a key driver of national competitiveness and security. In an era defined by rapid technological advancement and industrial transformation, nations strive to lead in the international science and technology arena, taking advantage of AI to address challenges across sectors such as agriculture, astronomy, and cybersecurity. AI’s role in enhancing productivity, sustainability, and security highlights its strategic importance, underscoring the urgency for countries to actively pursue AI development to secure a competitive edge in a globalised world. Methodology. The study employs a multi-faceted methodological approach. First, a comprehensive literature review and analysis of AI applications in various sectors, including agriculture, astronomy, and cybersecurity, is conducted to provide context on current advancements and trends. Secondly, a comparative analysis examines the strategic AI policies of leading nations to assess how different countries are positioning AI within their national agendas. Third, case studies of AI implementation in specific sectors, such as precision agriculture and cybersecurity, illustrate the practical impacts and potential benefits of a society-oriented approach to AI. The aim of this study is to analyse the strategic value of fostering an AI-driven society as a means of enhancing national competitiveness and securing leadership in international technological innovation. It aims to explore how AI can be harnessed to support sustainable development, improve sectoral efficiency, and protect against security threats, thus contributing to the overall socio-economic resilience and global standing of a nation. Results. The study reveals that the integration of AI across diverse sectors has led to significant efficiency gains, particularly in resource management, sustainability, and security. AI-driven advancements in agriculture, such as precision farming, contribute to higher productivity and environmental sustainability, while applications in astronomy support large-scale data processing for deep space exploration. In cybersecurity, AI has proven instrumental in identifying and countering cyber threats in real time. These findings confirm that an AI-centric societal model can enhance national resilience, drive economic stability, and bolster a country’s competitive position on the global stage. Conclusion. The emergence and development of the “artificial intelligence society” in the context of the technological transformation of the world is a process in which societies adapt to the profound changes caused by the introduction and development of artificial intelligence (AI) technologies. This process includes several key aspects: economic change, based on the automation of production processes, leading to increased efficiency and productivity; the creation of new markets and business models based on AI capabilities, including the redistribution of jobs and changes in employee skill requirements; social change, which is based on changing the way people interact with each other and with technology; transition to smart cities and communities, where AI helps to manage resources and ensure the comfort of life; impact on education, health and other areas of life through the introduction of personalised AI-based solutions; cultural changes aimed at transforming the values and worldview associated with AI technologies; emergence of new cultural practices and media formats based on AI; development of digital culture and its impact on traditional cultural forms; political and ethical challenges, including defining new regulatory and legal frameworks for AI regulation; ensuring ethical use of AI, avoiding discrimination and ensuring fairness; managing risks related to data security and privacy; technological development, based on the continuous improvement of AI algorithms and models; integration of AI into various sectors of the economy and everyday life’; development of infrastructure to support the large-scale implementation of AI (e.g. 5G networks, data centres); This process reflects the overall technological transformation of the world, where AI is becoming an integral part of economic, social, cultural, and political life. Key words: artificial intelligence, philosophy of society, national competitiveness, strategy.

Water resource systems display complex behavior that challenges our ability to identify paths toward improved management. Such behavior can arise from unanticipated feedbacks between social, ecological, and technological components that are conventionally studied and managed in disciplinary silos, often with limited consideration of interactions across scales of space and time. Convergence research driven by deep integration and co-production of knowledge within research teams is needed to better anticipate water resource system behavior and identify new approaches. We developed and applied a new framework—the Collaborative, Adaptive, and Multi-Scale (CAMS) systems thinking framework—to build a convergence research team around the task of characterizing a watershed as a complex system and hypothesize associated water management dynamics. The CAMS framework applies systems thinking methods within a broader integrated approach to engage and synthesize the knowledge and interests of an intellectually diverse research team and model a water resource system across spatial and temporal scales. Our case study of the Santa Fe Watershed in New Mexico reflects challenges and opportunities to manage water in the western United States of America. The specific methods applied within the framework included a six-session workshop on systems thinking, conceptual model development exercises with a longer-term subgroup, a structural analysis of system variables, and classroom-based projects. We discuss the successes, limitations, and potential of each method and how they interacted within the CAMS framework. We found that use of multiple systems thinking methods within the open-ended, iterative design of the framework provided a structure for long-term use that integrates disparate ideas, hypotheses, and findings from water sustainability research. Creating an inclusive environment within the research team was critical to the framework’s successful application and will remain a core consideration for ongoing work aimed at broader participation.

Summary of Defense Advanced Research Projects Agency's (DARPA) explainable artificial intelligence (XAI) program from the program managers' and evaluator's perspective. Defense Advanced Research Projects Agency (DARPA) formulated the explainable artificial intelligence (XAI) program in 2015 with the goal to enable end users to better understand, trust, and effectively manage artificially intelligent systems. In 2017, the 4-year XAI research program began. Now, as XAI comes to an end in 2021, it is time to reflect on what succeeded, what failed, and what was learned. This article summarizes the goals, organization, and research progress of the XAI program. Dramatic success in machine learning has created an explosion of new AI capabilities. Continued advances promise to produce autonomous systems that perceive, learn, decide, and act on their own. These systems offer tremendous benefits, but their effectiveness will be limited by the machine's inability to explain its decisions and actions to human users. This issue is especially important for the United States Department of Defense (DoD), which faces challenges that require the development of more intelligent, autonomous, and reliable systems. XAI will be essential for users to understand, appropriately trust, and effectively manage this emerging generation of artificially intelligent partners. The problem of explainability is, to some extent, the result of AI's success. In the early days of AI, the predominant reasoning methods were logical and symbolic. These early systems reasoned by performing some form of logical inference on (somewhat) human readable symbols. Early systems could generate a trace of their inference steps, which could then become the basis for explanation. As a result, there was significant work on how to make these systems explainable.1-5 Yet, these early AI systems were ineffective; they proved too expensive to build and too brittle against the complexities of the real world. Success in AI came as researchers developed new machine learning techniques that could construct models of the world using their own internal representations (eg, support vectors, random forests, probabilistic models, and neural networks). These new models were much more effective, but necessarily more opaque and less explainable. The year 2015 was an inflection point in the need for XAI. Data analytics and machine learning had just experienced a decade of rapid progress.6 The deep learning revolution had just begun, following the breakthrough ImageNet demonstration in 2012.6, 7 The popular press was alive with animated speculation about Superintelligence8 and the coming AI Apocalypse.9, 10 Everyone wanted to know how to understand, trust, and manage these mysterious, seemingly inscrutable, AI systems. 2015 also saw the emergence of initial ideas for providing explainability. Some researchers were exploring deep learning techniques, such as the use of deconvolutional networks to visualize the layers of convolutional networks.11 Other researchers were pursuing techniques to learn more interpretable models, such as Bayesian Rule Lists.12 Others were developing model-agnostic techniques that could experiment with a machine learning model—as a black box—to infer an approximate, explainable model, such as LIME.13 Yet, others were evaluating the psychological and human-computer interaction aspects of the explanation interface.13, 14 DARPA spent a year surveying researchers, analyzing possible research strategies, and formulating the goals and structure of the program. In August 2016, DARPA released DARPA-BAA-16-53 to call for proposals. The stated goal of explainable artificial intelligence (XAI) was to create a suite of new or modified machine learning techniques that produce explainable models that, when combined with effective explanation techniques, enable end users to understand, appropriately trust, and effectively manage the emerging generation of AI systems. The target of XAI was an end user who depends on decisions or recommendations produced by an AI system, or actions taken by it, and therefore needs to understand the system's rationale. For example, an intelligence analyst who receives recommendations from a big data analytics system needs to understand why it recommended certain activity for further investigation. Similarly, an operator who tasks an autonomous system needs to understand the system's decision-making model to appropriately use it in future missions. The XAI concept was to provide users with explanations that enable them to understand the system's overall strengths and weaknesses; convey an understanding of how it will behave in future/different situations; and perhaps permit users to correct the system's mistakes. The XAI program assumed an inherent tension between machine learning performance (eg, predictive accuracy) and explainability, a concern that was consistent with the research results at the time. Often the highest performing methods (eg, deep learning) were the least explainable and the most explainable (eg, decision trees) were the least accurate. The program hoped to create a portfolio of new machine learning and explanation techniques to provide future practitioners with a wider range of design options covering the performance-explainability trade space. If an application required higher performance, the XAI portfolio would include more explainable, high-performing, deep learning techniques. If an application required more explainability, XAI would include higher performing, interpretable models. The program was organized into three major technical areas (TAs), as illustrated in Figure 1: (a) the development of new XAI machine learning and explanation techniques for generating effective explanations; (b) understanding the psychology of explanation by summarizing, extending and applying psychological theories of explanation; and (c) evaluation of the new XAI techniques in two challenge problem areas: data analytics and autonomy. The original program schedule consisted of two phases: phase 1, Technology Demonstrations (18 months); and phase 2, Comparative Evaluations (30 months). During phase 1, developers were asked to demonstrate their technology against their own test problems. During phase 2, the original plan was to have developers test their technology against one of two common problems (Figure 2) defined by the government evaluator. At the end of phase 2, the developers were expected to contribute prototype software to an open source XAI toolkit. In May 2017, XAI development began. Eleven research teams were selected to develop the Explainable Learners (TA1) and one team was selected to develop the Psychological Models of Explanation. Evaluation was provided by the Naval Research Lab. The following summarizes those developments and the final state of this work at the end of the program. An interim summary of the XAI developments at the end of 2018 is given in Gunning and Aha.15 The program anticipated that researchers would examine the training process, model representations, and, importantly, explanation interfaces. Three general approaches were envisioned for model representations. Interpretable model approaches would seek to develop ML models that were inherently more explainable and more introspectable for machine learning experts. Deep explanation approaches would leverage deep learning or hybrid deep learning approaches to produce explanations in addition to predictions. Finally, model induction techniques would create approximate explainable models from more opaque, black-box models. Explanation interfaces were expected to be a critical element of XAI, connecting a user to the model to enable them to understand and interact with the decision making process. As the research progressed, 11 XAI teams explored a number of machine learning approaches, such as tractable probabilistic models16 and causal models and explanation techniques such as state machines generated by reinforcement learning algorithms,17 Bayesian teaching,18 visual saliency maps,19-24 and network and GAN dissection.24-26 Perhaps the most challenging and most unique contributions came from the combination of machine learning and explanation techniques27 to conduct well-designed psychological experiments to evaluate explanation effectiveness.28-31 As the program progressed, we also gained a more refined understanding of the spectrum of users and development timeline (Figure 3). The program structure anticipated the need for a grounded psychological understanding of explanation. One team was selected to summarize current psychological theories of explanation to assist the XAI developers and the evaluation team. This work began with an extensive literature survey on the psychology of explanation and previous work on explainability in AI.32 Originally, this team was asked to (a) produce a summary of current theories of explanation, (b) develop a computational model of explanation from those theories, and (c) validate the computational model against the evaluation results from the XAI developers. Developing computational models proved to be a bridge too far, but the team did gain a deep understanding of the area and successfully produced descriptive models. These descriptive models were critical to supporting the effective evaluation approaches, which involved carefully designed user studies, carried out in accordance with DoD human subject research guidelines. Figure 4 illustrates a top-level descriptive model of the XAI explanation process. Evaluation was originally envisioned to be based on a common set of problems, within the data analytics and autonomy domains. However, it quickly became clear that it would be more valuable to explore a variety of approaches across a breadth of problem domains. In order to evaluate the performance in the final year of the program, the evaluation team, led by Eric Vorm, PhD, of the US Naval Research Laboratory (NRL), developed an explanation scoring system (ESS). This scoring system provided a quantitative mechanism for assessing the designs of XAI user studies in terms of technical and methodological appropriateness and robustness. The ESS enabled the assessments of multiple elements of each user study, including the task, domain, explanations, explanation interface, users, hypothesis, data collection, and analysis to ensure that each study met the high standards of human subject research. XAI evaluation measures are shown in Figure 5, and include functional measures, learning performance measures, and explanation effectiveness measures. The DARPA XAI program demonstrated definitively the importance of carefully designing user studies in order to accurately evaluate the effectiveness of explanations in ways that directly enhance appropriate use and trust by human users, and appropriately support human-machine teaming. Often times, multiple types of measures (ie, performance, functionality, and explanation effectiveness) will be necessary to evaluate the performance of an XAI algorithm. XAI user study design can be tricky and the DARPA XAI program discovered that the most effective research teams were ones that featured diverse teams with cross-disciplinary expertise (ie, computer science combined with human-computer interaction and/or experimental psychology, etc.). The XAI program explored many approaches, as shown in Table 1. Interactive debugger interface for visualizing poisoned training datasets. Work is applied on the IARPA TrojAI dataset.33 Establishing objective/quantitative criteria to assess value of explanations for ML models34 CNN-based one-shot detector, using network dissection to identify the most salient features41 Explanations produced by heat maps and text explanations42 Human-machine common ground modeling Indoor navigation with a robot (in collaboration with GA Tech) Video Q&A Human-assisted one-shot classification system by identifying the most salient features Three major evaluations were conducted during the program: one during phase 1 and two during phase 2. In order to evaluate the effectiveness of XAI techniques, researchers on the program designed and executed user studies. User studies are still the gold standard for assessing explanations. There were approximately 12 700 participants in user studies carried out by XAI researchers, including approximately 1900 supervised participants, where the individual was guided through the experiment by the research team (eg, in person or on Zoom) and 10 800 unsupervised participants, where the individual self-guided through the experiment and was not actively guided by the research team (eg, Amazon Mechanical Turk). In accordance with policy for all US DoD funded human subjects research, each research protocol was reviewed by a local Institutional Review Board (IRB) and then a DoD human research protection office reviewed the protocol and the local IRB findings. As mentioned earlier, there seemed to be a natural tension between learning performance and explainability. However, throughout the course of the program, we found evidence that explainability can improve performance. From an intuitive perspective, training a system to produce explanations provides additional supervision, via additional loss functions, training data, or other mechanisms, that encourages a system to learn more effective representations of the world. While this may not be true in all cases and significant work remains to characterize when explainable techniques will be more performant, it provides hope that future XAI systems can be more performant than current systems while meeting user needs for explanations. There currently is no universal solution to XAI. As discussed earlier, different user types require different types of explanations. This is no different from what we face interacting with other humans. Consider, for example, a doctor needing to explain a diagnosis to a fellow doctor, a patient, or a medical review board. Perhaps future XAI systems will be able to automatically calibrate and communicate explanations to a specific user within a large range of user types, but that is still significantly beyond the current state of the art. One of the challenges in developing XAI is measuring the effectiveness of an explanation. DARPA's XAI effort has helped develop foundational technology in this area, but much more needs to be done, including drawing more from the human factors and psychology communities. Measures of explanation effectiveness need to be well established, well understood, and easily implemented by the developer community in order for effective explanations to become a core capability of ML systems. UC Berkeley's result21 demonstrating that advisability, the ability for an AI system to take advice from a user, improves user trust beyond explanations is intriguing. Certainly, users will likely prefer systems where they can quickly correct the behavior of a system in the same ways that humans can provide feedback to each other. Such advisable AI systems that can both produce and consume explanations will be key to enabling closer collaborations between humans and AI systems. Close collaboration is required across multiple disciplines including computer science, machine learning, artificial intelligence, human factors, and psychology, among others, in order to effectively develop XAI techniques. This can be particularly challenging, as researchers tend to focus on a single domain and often need to be pushed to work across domains. Perhaps in the future a XAI-specific research discipline will be created at the intersection of multiple current disciplines. Toward this end, we have worked to create an XAI Toolkit, which collects the various program artifacts (eg, code, papers, reports, etc.) and lessons learned from the 4-year DARPA XAI program into a central, publicly accessible location (https://xaitk.org/).48 We believe the toolkit will be of broad interest to anyone who deploys AI capabilities in operational settings and needs to validate, characterize, and trust AI performance across a wide range of real-world conditions and application areas. Today, we have a more nuanced, less dramatic, and, perhaps, more accurate understanding of AI than we had in 2015. We certainly have a more accurate understanding of the possibilities and the limitations of deep learning. The AI apocalypse has faded from an imminent danger to a distant curiosity. Similarly, the XAI program has produced a more nuanced, less dramatic, and, perhaps, more accurate understanding of XAI. The program certainly acted as a catalyst to stimulate XAI research (both inside and outside of the program). The results have produced a more nuanced understanding of XAI uses and users, the psychology of XAI, the challenges of measuring explanation effectiveness, as well as producing a new portfolio of XAI ML and HCI techniques. There is certainly more work to be done, especially as new AI techniques are developed that will continue to need explanation. XAI will continue as an active research area for some time. The authors believe that the XAI program has made a significant contribution by providing the foundation to launch that endeavor. David Gunning (now retired) is a three-time DARPA program manager, who created and managed the XAI program from its inception in 2016 to its mid-point in 2019. His portfolio of DARPA research programs made significant contributions to the development of AI over the past 25 years. He led the Personalized Assistant that Learns (PAL) program, which produced the technologies behind Apple's Siri. His Command Post of the Future (CPoF) program was later adopted by the US Army as their Command and Control system for use during the Iraq and Afghanistan conflicts. Between DARPA tours, David served in senior positions at Facebook AI, Palo Alto Research Center, Vulcan Inc, Cycorp and co-founded SET Corp. Eric Vorm, PhD, is a cognitive systems engineer and serves as the Deputy Director for the Laboratory for Autonomous Systems Research at the US Naval Research Laboratory in Washington, DC. Dr Vorm led the evaluation team for the DARPA Explainable AI program, and led the development of the first comprehensive criteria for the evaluation of explanations generated by machine learning. Dr Vorm's research focuses on the design of intelligent systems to achieve ideal human-machine teaming, with special emphasis on the role of transparency and explainability in supporting appropriate trust, safety, and reliability in high-risk, time-sensitive operational domains. Jennifer Yunyan Wang, PhD, is a computational neuroscientist with a special focus on AI. As Systems, Engineering and technical Assistance (SETA) contractor to DARPA, she provided technical support and expertise to several programs including XAI, L2M, GARD, and AIE RED. After finishing postdoctoral fellowships in experimental neuroscience at Johns Hopkins University and the Food and Drug Administration, Jennifer joined Quantitative Scientific Solutions in 2018 as a consultant for government R&D and think tanks including IARPA and Center for Security and Emerging Technology. Matt Turek, PhD, joined DARPA's Information Innovation Office (I2O) as a program manager in July 2018 and took over as program manager of the XAI program in 2019. His portfolio also includes the Media Forensics (MediFor), Semantic Forensics (SemaFor), and Machine Common Sense (MCS) programs, as well as the Reverse Engineering of Deceptions (RED) AI Exploration. His research interests include computer vision, machine learning, artificial intelligence, and their application to problems with significant societal impact. Prior to his position at DARPA, Turek led a team at Kitware Inc developing computer vision technologies including large scale behavior recognition and modeling, object detection and tracking, activity recognition, normalcy modeling, and anomaly detection. Data sharing is not applicable to this article as no new data were created or analyzed in this editorial.

The transition from traditional IT systems to Artificial Intelligence (AI) solutions represents a transformational shift in the technological landscape, requiring new paradigms in change management. This paper explores how lessons learned from large-scale IT integrations can inform effective strategies for AI implementation. We begin with an examination of change management principles, focusing on communication strategies, training and skill development, and phased implementation approaches in IT. The unique challenges of AI, including its complexity, rapid advancement, and the shift to data-driven decision-making, are analysed to understand the adaptation needed in change management strategies. A hypothetical case study of a bank's AI adoption demonstrates the application of these principles, highlighting results and key takeaways. The paper culminates in best practices for AI-era change management, emphasizing innovation, cross-functional teams, ethical frameworks, and impact measurement. This comprehensive analysis underscores the enduring importance of change management in technological transformations, offering a call to action for organizations to pro actively embrace and adapt these strategies. The transition from traditional IT systems to Artificial Intelligence (AI) solutions represents a informative shift in the technological landscape, requiring new paradigms in change management. This paper explores how lessons learned from large-scale IT integrations can inform effective strategies for AI implementation. As organizations move towards AI-driven solutions, understanding the nuances of change management becomes crucial to harnessing Al's full potential. We begin with an examination of foundational change management principles, focusing on critical components such as communication strategies, training and skill development, and phased implementation approaches, traditionally used in IT integrations. These aspects are pivotal in creating a structured environment where AI can thrive and integrate seamlessly with existing systems. The unique challenges of AI implementation are then scrutinized, highlighting its inherent complexity, rapid technological advancement, and the significant shift to data-driven decision-making processes. AI systems often operate in ways that are not immediately transparent, requiring organizations to navigate the uncertainties and ethical considerations inherent in AI deployment. This analysis provides a comprehensive understanding of the adaptations needed in change management strategies to accommodate these challenges. It explores how organizations can leverage AI to transform operations while addressing the potential risks associated with its adoption. To illustrate the practical application of these principles, a hypothetical case study of a bank's AI adoption is presented. This scenario offers an in-depth look at how strategic change management approaches can be tailored to support AI integration, providing insights into successful practices and potential pitfalls. The case study highlights specific outcomes, key takeaways, and lessons learned from aligning AI initiatives with organizational goals and stakeholder expectations. The paper culminates in outlining best practices for AI-era change management, emphasizing the importance of fostering a culture of innovation, developing cross-functional teams, and establishing ethical AI frameworks. It stresses the need for organizations to develop robust methods for measuring and communicating the impact of AI technologies, ensuring that they contribute positively to business objectives and societal values. These best practices serve as a guide for organizations looking to navigate the complexities of AI integration effectively. This comprehensive analysis underscores the enduring importance of change management in technological transformations, particularly as we enter an era dominated by AI advancements. It offers a call to action for organizations to pro-actively embrace and adapt these strategies, recognizing that effective change management is not just a facilitator of technological adoption but a catalyst for innovation and growth. By understanding and implementing these tailored strategies, organizations can position themselves at the forefront of technological evolution, leverage AI to drive meaningful change and sustainable success.

The article "Embracing Real AI: A Call to Action for Medical Physicists in Healthcare" urges medical physicists to prepare for the integration of artificial intelligence (AI) into healthcare practices, emphasizing their pivotal role in adapting to technological advancements. The authors advocate for embracing AI through advocacy, broadening perspectives, and enhancing coordination and communication. They propose an ABC strategy focusing on increasing educational initiatives, fostering interdisciplinary collaboration, and creating team collaboration to facilitate AI integration. The commentary highlights AI's potential in enhancing diagnostics, personalizing medicine, and automating routine tasks while addressing challenges such as data sharing and the role of federated learning. The article calls for medical physicists to lead in embracing AI, emphasizing continuous learning and collaboration to leverage its potential for improving healthcare and patient care. Medical physicists have consistently demonstrated strong interest in developing proficiency in the adoption of new technological advancements. The roots of the profession come from the radiation sciences, including radiation protection, radiation therapy, diagnostic imaging, and nuclear medicine.1 As science and technology continued to evolve, medical physicists' roles have extended into other non-radiation domains, such as non-ionizing-radiation-based imaging (ultrasound and magnetic resonance), molecular imaging, computer aided diagnosis (CAD), information technologies, and data science.2 In addition, medical physicists gradually have adopted increasingly more active roles in ensuring the professional education of other radiology/radiation oncology team members, maintaining high quality standards via quality assurance (QA) methods. They also play a major role in advising the hospital management on medical devices and software acquisition. The continuing expansion of these roles and responsibilities has put medical physicists on the forefront of embracing emerging technologies, making the profession one of the most technical and versatile in healthcare settings. Currently, as our field grows in importance, we medical physicists seek to continue to engage in significant ways to for increased contributions and roles in human health. This commentary/opinion urges medical physicists to prepare for their expanding roles in the field of AI and its implementation and oversight in clinical practice. Medical physicists must embrace "Real AI" to help integrate AI into healthcare practices. Conceptually we advocate for a strategy that involves Real AI through advocacy, broadening, and enhancing coordination/communication (an ABC strategy). In our current and future work medical physicists will use AI to automate routine tasks, allowing medical physicists to focus on more complex tasks. Furthermore, Medical Physics will use AI to enhance efficiency, safety, diagnostic and therapeutic applications, and for personalized medicine. However, as we have done in the past with other complex concepts (such as radiation), medical physicists need to be prepared for the potential risks and ethical dilemmas associated with AI, such as bias and lack of transparency. It will be important that Medical Physicists prepare for the rapidly changing AI landscape, and continue learning, gain hands-on experience, and collaborate with other AI experts in the healthcare environment. This paper aligns with the already approved guidance document developed by the AAPM in conjunction with International Atomic Energy Agency (IAEA)3 that discusses how medical physicists can ensure the effective implementation and management of AI systems. It is crucial for the Clinical Quality Management Program (CQMP) personnel to receive regular training and updates on relevant guidelines and legislation. Clear communication channels should be established with IT experts, vendors, and other stakeholders for smooth coordination.4 Comprehensive documentation should be developed to ensure compliance with contractual obligations and guidelines. The clinical team should be involved in acceptance testing and discussions, depending on the clinical purpose of the AI system.4 Protocols for data collection and curation should be established, along with the development of standardized validation datasets for performance evaluation.4 A system for monitoring updates to AI systems and models should be implemented, with the CQMP leading new acceptance/commissioning rounds for any updates. Lastly, mechanisms for continuous evaluation and improvement of the CQMP processes should be established, which could involve regular audits, feedback mechanisms from end-users, and incorporating lessons learned from previous rounds.4 Nowadays, major healthcare systems in the US consider their data as immensely valuable assets that require rigorous protection to ensure Health Insurance Portability and Accountability Act (HIPAA) compliance, as well as intellectual property considerations. It can be very difficult for researchers to share clinical data with vendors for development purposes without a significant return being specified to the institution, such as joint intellectual property or substantial grant funding. Instead, these healthcare systems encourage their researchers to commercialize their findings independently, allowing the institution to retain full rights to intellectual property. That said, the realization of federated learning would be a significant advancement. To achieve this, a powerful pre-trained model that would be adaptable to operation on different scales and in various clinical scenarios is necessary. It is plausible that local adaptation may not require substantial computing power or AI expertise. This concept is particularly intriguing and could be beneficial to smaller centers and clinics in underserved areas. However, the primary challenge is the cost. As we become more reliant on AI systems like OpenAI's ChatGPT or Google Gemini, we often overlook the fact that these conveniences come with a hefty price tag, costing billions of dollars to develop and maintain.5 As medical physicists we and other healthcare professionals can anticipate that AI will significantly transform healthcare, improving efficiency, accuracy, and the level of detail that can be extracted from imaging, and methods of therapy. These technological advancements are expected to bring immense value to the field, offering a new horizon in diagnostic and therapeutic capabilities. Yet, we also must recognize that it also introduces potential significant risks and ethical dilemmas. One of the primary concerns is the possibility of bias in AI, which can stem from the training data, the algorithms, or their application, leading to potentially detrimental effects on patient care. As medical physicists, we should acknowledge that the complexity and lack of transparency in AI decision-making processes present obstacles in terms of accountability and rectifying errors and requires greater oversight and responsibility. The integration of AI also has great capacity in redefining the role of medical physicists, impacting education and employment within the field. Addressing these issues necessitates the creation of ethical standards for AI in healthcare, emphasizing transparency, responsibility, and equity, with contributions from diverse stakeholders, including patients, medical professionals, and ethicists.6 Such measures are crucial to ensure the responsible utilization of AI in healthcare, and ultimately serve the best interests of patients and society. We anticipate that continued guidance from our professional societies will be helpful as our collective communities develop methods and approaches that help us learn, adopt, and employ AI responsibly. Advocacy: increase educational initiative, public awareness, and recommending processes at all levels of the clinical workforce, as well as patient engagement. Broadening Perspectives: encourage Interdisciplinary Collaborations that allow medical physicists to work with professionals from other disciplines such as computer science, data science, and biomedical engineering, to gain insights into different perspectives on AI applications in healthcare. This enables medical physicists to provide continuing education and connect the community with research opportunities. Improving Coordination and Communication through creating team collaboration: enhance communication with healthcare professionals, administrators, and patients by clearly defining and articulating the role of medical physicists in AI applications. Promote the sharing of knowledge, as exemplified by creating data repositories through contributions, to further creating the foundation of our understanding and application of AI in the field. We consider the concept of Real AI in our context to be aimed at providing and/or qualifying a ready AI product that has undergone a rigorous QA process, that is free of false additives and biases, with data carefully curated to represent the demographics and be attuned to the needs of the clinic, sourced with proper ingredients, and abiding by laws and regulations that can ensure the product serves the common health needs of patients and benefits the public's interest. What AI 'is' and what it 'is not' is a complex topic that warrants further exploration and understanding, but one vital for comprehension of what utility AI can fulfill in the clinical process, what its advantages and limitations are, and how it can be curated to perform in the clinical scenarios relevant to a particular radiology/radiation oncology practice. Multiple data-analysis algorithms have been created over the course of years, and not all of them qualify as AI.7 What distinction(s) lie in what constitutes AI? One possible interpretation is that AI is a system that can adapt to new data, or a system that generates insights driven by data. AI systems are designed to "learn" and adapt to new data and be stable over the course of introducing data perturbations or employ model adaptation mechanisms. AI systems can adjust the underlying data-processing mechanisms based on the input they receive, which allows them to improve their performance and make more accurate predictions or decisions over time. This is often achieved through techniques such as machine learning, where algorithms are trained on a dataset and then used to make predictions or decisions without being explicitly programed to perform the task.8 Understanding how such datasets are selected, what data needs to be fed into AI model to achieve desired results, and how to prevent common pitfalls and ethical conundrums associated with the use of AI models requires additional training that might yet be lacking in the traditional training of the radiology/radiation oncology adjacent specialists. The scope of involvement of each member of the team when it comes to AI integration into the clinic continues to be determined as the field rapidly evolves. When it comes to the role of medical physicists in conjunction with AI, an open discussion of the exact responsibilities is still ongoing, and feedback is encouraged from all the members of the community. So, what can medical physicists do? They can use AI to enhance quality improvement and safety by analyzing medical data to identify trends, patterns, and outliers.9 This can lead to the identification of areas for improvement or potential safety hazards and help them enter the realm of Responsible AI. AI can also improve diagnostic and therapeutic techniques by enhancing the quality of medical imaging and automating image interpretation.10 Furthermore, AI can help in integrating diagnostics, personalized medicine, and theragnostics by analyzing large datasets to tailor treatment plans to individual patients.11 This can lead to more effective and personalized care. AI can also automate routine tasks in medical physics, such as treatment planning and QA processes, leading to increased efficiency.12 Lastly, AI techniques like machine learning and deep learning can be leveraged for research and development to analyze complex datasets, discover patterns, and develop innovative techniques for disease detection, treatment, and monitoring.13 Whether it involves developing AI-driven solutions like automated segmentation, dose calculations, addressing intricate problems in the clinic, or potentially even contributing to open-source AI initiatives, such activities will empower medical physicists to enhance their skills and make tangible contributions to the advancement of healthcare. Embracing AI not only fosters a sense of accomplishment but also opens doors to the world of `automation' and scaling that will pervade all technologies of the future. The AHAIBC committee is at the center of bringing the medical physicist forward by developing curriculum concepts, bootcamps, and engendering engagement for our society. Integration of AI into the realm of medical physics education is critical, especially considering the potential significance of incorrect AI usage or misapplication. The physicist is responsible for installing and commissioning the AI software, ensuring the modeling is not biased, performing continuing QA on the hospital data and processes, and establishing efficient resource management. Embracing education in AI offers new benefits for medical physicists as it is already revolutionizing various industries and professional practices and we need to be equally prepared. One way to engage and prepare healthcare professionals for the upcoming AI wave is to start with the roots of quality safety and assurance. To do this, we should enable a comprehensive QA program that encompasses all clinical operations related to medical fields including radiology, nuclear medicine, and radiation oncology. Ensuring the safe operation of hardware, software, clinical operation processes and machinery is of utmost importance and one of the most crucial responsibilities of a medical physicist. A Real AI approach can be highly beneficial in achieving the goal of safe clinical implementation. Understanding the potential and limitations of AI serves as a cornerstone for fostering engagement not only within our profession but with other healthcare providers. Continuous learning and participation in hands-on experience are essential components for navigating the complexities of AI applications within healthcare. Collaboration, networking, and exploring AI's purpose and impact are equally vital in this journey. Additionally, some physicists may choose personal projects, embracing challenges in small groups, and actively contributing to AI-focused teams to amplify the motivation and expertise of our field. Insights through personal and collaborative opportunities ultimately provide for and encourage professional growth and innovation within our medical physics field. Some medical physicists may be able to attend specialty meetings and conferences dedicated to AI which further enriches their knowledge base and provides them avenues for fruitful collaboration. There are successful educational programs such as the Radiological Society of North America Artificial Intelligence (RSNA AI)-certificate program.14 Interdisciplinary cooperation and inter-institutional collaboration for AI experts is of paramount importance for integrating AI into medical physicists' practice on a larger scale, and mechanisms enabling this collaboration should be provided to the community. In summary, the authors believe that being prepared for and embracing the changes that AI is already bringing at the current time will benefit our community, healthcare, patient care, and society at large immediately and for the future. We are at a critical juncture, which can be considered a fourth industrial revolution, where AI and automation are applied more broadly. Medical physicists have a pivotal role to play in this revolution. We need to position ourselves at the forefront of 'Real AI' and lead the charge in this exciting new era. It is time for action, and we can take the first steps with potentially just a few ABCs. All authors contributed their efforts in writing and editing this call for action. ChatGPT search engine has been utilized to provide additional background to the subject of matter for illustrative purposes. The authors appreciate members of the Ad. The authors declare no conflicts of interest. The content for this call for action has been edited with the help of large language models ChatGPT and Google NotebookLM.

During the summer of 2008 in Washington, DC, the Cognitive Science Society celebratedthe 30th anniversary of its seminal 1979 conference in San Diego. The 2008 conferenceorganizers—Bradley Love, Ken McRae, and Vladimir Sloutsky—commissioned a sympo-sium to celebrate the occasion. In discussing possibilities, we agreed that the symposiumshould not simply address the Society’s origins and subsequent history, but that it shouldfocus on contributions from the disciplines and theoretical perspectives central to CognitiveScience, along with their future directions.We originally settled on five disciplines and five theoretical perspectives, and then weinvited 10 active established researchers to address them at the conference. To accommo-date these 10 speakers, two symposia were presented, one on disciplines and one onperspectives. Each speaker was asked to address: (a) What was your discipline⁄perspectivelike at the time of the 1979 conference? (b) How has the discipline⁄perspective changedover the past 30 years to what it is today? (c) How do you foresee the discipline⁄perspectivechanging in the next 30 years?Because of time constraints, we could not include all disciplines and perspectives centralto Cognitive Science. Fortunately, however, we were able to remedy this limitation byasking additional researchers to contribute articles here. The resulting collection of articlescovers disciplines and perspectives that have been central to Cognitive Science for the past30 years and that are likely to be central for the next 30 years and beyond. Specifically, thedisciplines (and the authors addressing them) include the following:Psychology (Dedre Gentner)Artificial Intelligence (Kenneth D. Forbus)Philosophy (William Bechtel)Linguistics (Elissa L. Newport)

ABSTRACT: The synergistic evolution of Artificial Intelligence (AI) and Data Science represents one of the most transformative technological developments of the 21st century. This paper provides a comprehensive survey and analysis of contemporary advances across this integrated field, moving beyond siloed examinations to explore their convergence. We trace the foundational journey from classical machine learning (ML) to the deep learning revolution and onward to today's frontier of generative AI, large language models (LLMs), and neuro-symbolic systems. Crucially, we position these algorithmic breakthroughs within the enabling ecosystem of modern data science, emphasizing the critical role of advances in data engineering (e.g., vector databases, data lakes), computational frameworks (e.g., distributed computing, specialized hardware like TPUs/GPUs), and methodological rigor (e.g., MLOps, explainable AI - XAI). A novel framework, the "AI-Data Science Flywheel," is introduced to conceptualize how scalable data infrastructure fuels more sophisticated AI models, whose outputs in turn generate new, high-value data, creating a virtuous cycle of innovation. Our methodology combines a systematic literature review of over 200 key publications from 2018-2024 with a quantitative meta-analysis of performance benchmarks across 10 standard datasets (e.g., ImageNet, GLUE, Open LLM Leaderboard) and a qualitative case study analysis of three industry verticals: healthcare (AI-driven drug discovery), finance (algorithmic risk assessment), and climate science (climate informatics). Results indicate that while model capabilities on narrow tasks have seen near-superhuman performance gains (e.g., 99.5% accuracy on ImageNet classification), significant challenges persist in areas of generalization, energy efficiency, bias mitigation, and the integration of causal reasoning. The case studies reveal a common pattern: successful deployment hinges less on model novelty and more on robust data pipelines and effective human-AI collaboration frameworks. This paper concludes that the future trajectory of AI and Data Science will be defined by a shift from "bigger models" to "smarter, more efficient, and more responsible systems." Key research vectors include the pursuit of artificial general intelligence (AGI) pathways, federated learning for privacy preservation, quantum machine learning, and the development of comprehensive ethical and regulatory frameworks to ensure these powerful technologies yield broad societal benefit.

Tourism ranks among the world's largest industries, and its sustained expansion has paralleled swift advancements in technology. Artificial Intelligence (AI) is increasingly recognized as a transformative force in tourism, offering human-like capabilities that enhance decision-making and service automation. Its application across the sector improves operational efficiency and personalizes customer experiences, thereby fostering innovation and competitiveness. However, the rapid integration of AI also presents conceptual, theoretical, and societal challenges that require critical examination. The research aims to synthesize the conceptual and theoretical research on AI in tourism from 2019 onwards. It examines key themes, theoretical perspectives, methodological rigor, and research gaps in the existing literature. Further goal is to identify thematic areas with a specific focus on AI applications. The study followed the PRISMA guidelines to conduct a systematic literature review (SLR). Academic databases, including Scopus and Web of Science, were searched to identify scientific-relevant peer-reviewed articles. From an initial pool of over 400 studies, we identified 45 significant journal articles and selected them for an in-depth analysis, that collectively illuminate how AI is reshaping tourism research and practice. Studies have drawn on innovation diffusion theory to explain adoption patterns, technology acceptance models to gauge user and employee attitudes, and service quality and cocreation theories to understand how AI can add value to the customer experience. It also highlighted the evolution of AI research in tourism, from conceptual discussions to empirical investigations. Gaps and challenges in the research were identified, including a limited focus on human-AI interaction, ethical concerns, and methodological rigor. The review concludes that AI has the potential to transform tourism by enhancing efficiency, personalization, and sustainability. The findings reveal that AI has been envisioned as a catalyst for transformation in the tourism industry, with applications ranging from intelligent forecasting and revenue management to service automation via robots and hyper-personalized travel experiences. AI-driven analytics can improve decision support for revenue management, capacity planning, and marketing strategy. However, realizing this potential requires addressing the improvement of technological competence of human resources, ethical issues, and implementation strategies.

Background:Artificial Intelligence (AI) is rapidly transforming the healthcare industry by providing innovative solutions to enhance efficiency, accuracy, and patient outcomes. Its applications span multiple healthcare domains, including nursing, midwifery, radiology, pharmacy, health administration, and social work. In nursing, AI supports decision-making and workflow optimization, while in midwifery, it assists in maternal and fetal health monitoring. Radiology has benefited significantly from AI's ability to analysed medical imaging with high precision, and pharmacy has seen advancements in drug discovery and personalized medicine. Health administration utilizes AI to streamline operational processes, improve resource allocation, and support data-driven decision-making. In social work, AI aids in identifying social determinants of health, enabling targeted interventions and improving access to care. Despite its potential, integrating AI into healthcare presents challenges, including ethical concerns and data security. Aim:This systematic review aims to explore and analyze the role of AI in healthcare, focusing on its applications in nursing, midwifery, radiology, pharmacy, health administration, and social work. The study seeks to identify the benefits, challenges, and future opportunities for AI integration in these domains to enhance healthcare delivery. Methods:A systematic review of the literature was conducted using peer-reviewed journals and reputable databases. The search strategy focused on publications related to AI in healthcare, specifically in nursing, midwifery, radiology, pharmacy, health administration, and social work. Articles were selected based on their relevance, methodological rigor, and recent contributions to the field. Data were extracted and synthesized to identify key themes, benefits, and challenges associated with AI applications. Results:AI demonstrated significant potential across all domains. In nursing, it improved care delivery through predictive analytics and decision-support systems. Midwifery benefited from enhanced monitoring and early detection of complications in maternal and fetal health. Radiology saw breakthroughs in medical imaging interpretation, aiding early disease detection. In pharmacy, AI accelerated drug discovery and personalized treatment development. Health administration leveraged AI for resource management, predictive modeling, and process optimization. In social work, AI facilitated the analysis of complex social data, enabling effective interventions and improved care for vulnerable populations. Common challenges included ethical considerations, data privacy concerns, and the need for comprehensive regulatory frameworks. Conclusion:AI is revolutionizing healthcare, offering numerous benefits while also posing challenges that must be addressed for successful integration. By leveraging its capabilities responsibly, AI has the potential to transform healthcare delivery, improving outcomes and advancing the quality of care across nursing, midwifery, radiology, pharmacy, health administration, and social work.

Artificial Intelligence (AI) has emerged as a transformative enabler across strategic management, qualitative research, and crowdsourced operational systems. However, adoption is shaped by human judgement, organisational processes, and socio-technical factors. Existing literature often examines AI applications in isolation, overlooking integrative approaches that balance technical capability with human and ethical oversight. This study systematically synthesises evidence to examine AI’s impact across multiple domains, identifying patterns, limitations, and opportunities, and proposes a human-centred framework for responsible deployment. A systematic integrative review was conducted, encompassing peer-reviewed journals, technical reports, and policy documents. Data extraction focused on AI capabilities, human-AI interaction, governance, methodological rigour, and socio-technical integration. Thematic analysis identified recurring patterns and gaps across domains. This study reveals that AI-driven decision-support systems enhance predictive analytics, scenario planning, and resource allocation, yet require managerial expertise, governance, and interpretive oversight to translate insights into actionable strategy. Furthermore, AI-assisted tools improve thematic analysis, coding, and data synthesis efficiency, but human interpretation remains critical to maintain contextual depth, methodological rigour, and ethical integrity. Lastly, Platforms such as Waze and Google Maps demonstrate real-time operational value, yet outcomes are contingent on data quality, user engagement, and trust, highlighting the socio-technical dependencies of AI deployment. The Triadic AI Integration Framework (TAIF) operationalises these insights by linking AI capabilities, human interpretation, and organisational processes within a human-centred, ethically governed structure. Effective AI adoption requires interpretive oversight, socio-technical alignment, and cross-domain integration to maximise strategic, research, and operational impact. Future research should empirically test TAIF, explore socio-technical adaptation, and examine long-term organisational and societal outcomes.

The umbrella review by Karimzadhagh et al. is an ambitious and comprehensive attempt to distill the rapidly expanding literature on artificial intelligence (AI) in skin cancer diagnosis into meaningful clinical insight [1]. It succeeds in scope but simultaneously exposes the weaknesses that continue to plague AI research in dermatology. By synthesizing 11 meta-analyses covering over 100,000 lesions and more than a million images, the authors offer an impressive data-driven overview of AI's diagnostic capabilities across melanoma, basal cell carcinoma (BCC), and squamous cell carcinoma (SCC). Yet, beneath the encouraging numbers lies a more sobering truth: AI's performance remains context-dependent, uneven, and often overstated. The review's findings reinforce a familiar pattern: AI performs admirably on curated data sets but falters when tested against real-world diversity. Convolutional neural networks (CNNs) and support vector machines (SVMs) appear to dominate the field, achieving sensitivities and specificities in the mid-80% range when distinguishing melanoma among all lesion types, and even higher (above 90%) when confined to narrower diagnostic tasks. While these results look impressive on paper, they are far from the near-perfect accuracies often advertised in single-institution studies. The analysis on nonmelanoma skin cancers further illustrates this disparity: models integrating dermoscopy, optical coherence tomography (OCT), and smartphone images show high specificity for BCC, but hyperspectral approaches trade precision for sensitivity, yielding inflated performance that might not hold outside controlled environments. This variability underscores the uncomfortable gap between algorithmic success in the lab and clinical reliability at the bedside. One of the more striking and encouraging takeaways, however, is that AI seems to elevate nonspecialists far more than it assists experts. In comparative analyses, nurse practitioners and general practitioners gained the most from AI support, narrowing the performance gap with dermatologists. This is not just a technical finding; it is a potential paradigm shift. If implemented thoughtfully, AI could democratize dermatologic expertise, improving triage and early detection in primary care and telemedicine. But this promise is fragile. Without proper calibration and validation across diverse populations, AI risks amplifying disparities rather than reducing them [2]. The review's note that most data sets are biased toward fair-skinned individuals should set off alarms. It is disheartening that, even in 2025, equity remains an afterthought in AI model development. A parallel caution applies to large language models: although widely used by patients for self-triage and health advice, they frequently perform worse in real-world settings than on benchmark prompts, with hallucinations and overconfident errors that can mislead lay users [3, 4]. Methodological rigor also emerges as a serious concern. Only one meta-analysis met high AMSTAR-2 quality standards. Such weaknesses erode confidence in the aggregated findings and reflect a broader trend of quantity over quality in AI research. The authors rightly highlight that performance often declines when algorithms are tested on independent or public data sets, exposing the pervasive problem of overfitting. This is more than a technical flaw; it is an ethical one, given that overstated accuracy may mislead clinicians and patients alike. If AI is to move from journals to clinics, it must be subjected to the same level of scrutiny as any medical device, including transparent data sharing, external validation, and prospective trials in real-world settings. Despite these limitations, this umbrella review performs an important service. It tempers the inflated expectations surrounding AI and grounds the conversation in empirical evidence. As it stands, AI is not yet the dermatologist's equal, but it could become an indispensable ally if developed and validated responsibly. Karimzadhagh et al. remind us that technological sophistication cannot yet substitute for scientific integrity [1]. Until AI research in dermatology fully embraces methodological discipline and population diversity [5], its promise will remain aspirational rather than transformative. The authors declare no conflicts of interest. This article is linked to https://doi.org/10.1111/ijd.17981. Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.

Special issue on SoC and AI processors

Background: This narrative review examines the role of artificial intelligence (AI) in bowel sound analysis for the diagnosis and management of inflammatory bowel disease (IBD). Inflammatory bowel disease (IBD), encompassing Crohn’s disease and ulcerative colitis, presents a significant clinical burden due to its unpredictable course, variable symptomatology, and reliance on invasive procedures for diagnosis and disease monitoring. Despite advances in imaging and biomarkers, tools such as colonoscopy and fecal calprotectin remain costly, uncomfortable, and impractical for frequent or real-time assessment. Meanwhile, bowel sounds—an overlooked physiologic signal—reflect underlying gastrointestinal motility and inflammation but have historically lacked objective quantification. With recent advances in artificial intelligence (AI) and acoustic signal processing, there is growing interest in leveraging bowel sound analysis as a novel, non-invasive biomarker for detecting IBD, monitoring disease activity, and predicting disease flares. This approach holds the promise of continuous, low-cost, and patient-friendly monitoring, which could transform IBD management. Objectives: This narrative review assesses the clinical utility, methodological rigor, and potential future integration of artificial intelligence (AI)-driven bowel sound analysis in inflammatory bowel disease (IBD), with a focus on its potential as a non-invasive biomarker for disease activity, flare prediction, and differential diagnosis. Methods: This manuscript reviews the potential of AI-powered bowel sound analysis as a non-invasive tool for diagnosing, monitoring, and managing inflammatory bowel disease (IBD), including Crohn’s disease and ulcerative colitis. Traditional diagnostic methods, such as colonoscopy and biomarkers, are often invasive, costly, and impractical for real-time monitoring. The manuscript explores bowel sounds, which reflect gastrointestinal motility and inflammation, as an alternative biomarker by utilizing AI techniques like convolutional neural networks (CNNs), transformers, and gradient boosting. We analyze data on acoustic signal acquisition (e.g., smart T-shirts, smartphones), signal processing methodologies (e.g., MFCCs, spectrograms, empirical mode decomposition), and validation metrics (e.g., accuracy, F1 scores, AUC). Studies were assessed for clinical relevance, methodological rigor, and translational potential. Results: Across studies enrolling 16–100 participants, AI models achieved diagnostic accuracies of 88–96%, with AUCs ≥ 0.83 and F1 scores ranging from 0.71 to 0.85 for differentiating IBD from healthy controls and IBS. Transformer-based approaches (e.g., HuBERT, Wav2Vec 2.0) consistently outperformed CNNs and tabular models, yielding F1 scores of 80–85%, while gradient boosting on wearable multi-microphone recordings demonstrated robustness to background noise. Distinct acoustic signatures were identified, including prolonged sound-to-sound intervals in Crohn’s disease (mean 1232 ms vs. 511 ms in IBS) and high-pitched tinkling in stricturing phenotypes. Despite promising performance, current models remain below established biomarkers such as fecal calprotectin (~90% sensitivity for active disease), and generalizability is limited by small, heterogeneous cohorts and the absence of prospective validation. Conclusions: AI-powered bowel sound analysis represents a promising, non-invasive tool for IBD monitoring. However, widespread clinical integration requires standardized data acquisition protocols, large multi-center datasets with clinical correlates, explainable AI frameworks, and ethical data governance. Future directions include wearable-enabled remote monitoring platforms and multi-modal decision support systems integrating bowel sounds with biomarker and symptom data. This manuscript emphasizes the need for large-scale, multi-center studies, the development of explainable AI frameworks, and the integration of these tools within clinical workflows. Future directions include remote monitoring using wearables and multi-modal systems that combine bowel sounds with biomarkers and patient symptoms, aiming to transform IBD care into a more personalized and proactive model.

Lack of methodological rigor and limited coverage of generative artificial intelligence in existing artificial intelligence reporting guidelines: a scoping review.