Articles tagged with Artificial Intelligence
Current large language models struggle with early diagnostic reasoning and cannot be relied upon for patient-facing clinical decisions. The study highlights the need for further improvement in these areas to enhance patient care.
The new guidance provides practical routes forward for AI use in every publishing role, promoting accountability, transparency, risk-awareness, and innovation. It introduces structured checkpoints and ready-to-use tools to make responsible AI use practical, not theoretical.
A Drexel University study found that teens are concerned about their attachment to AI chatbots, using them for emotional support and entertainment. The study suggests that these interactions can evolve into addiction-like behavior, affecting their lives offline.
New research suggests that AI-powered price discrimination could lead to hidden, personalised pricing that consumers cannot see or understand, potentially reducing trust and affecting behavior. Regulators are now facing pressure to decide whether such practices cross the line under EU competition law.
Researchers developed a new training technique, HarmonyGNN, to improve the accuracy of graph neural networks in heterophilic graphs. The framework achieved state-of-the-art performance on four heterophilic graphs with accuracy improvements ranging from 1.27% to 9.6%.
A new study analyzed thousands of textual descriptions of two-person social interactions using generative AI, resulting in a comprehensive taxonomy of categories. The findings provide a data-driven framework for understanding the structure of social situations, including associations with conflict, power, and duty.
The study, led by Lucy Liu and L. Mahadevan, shows that adding the right amount of noise to individual robot movements can ease gridlock and improve efficiency in crowded environments. The researchers used computer simulations and experiments with small robots to test their ideas.
A custom-built AI system helped uncover how bacterial communities organize themselves, showing that early moments of a biological transition carry more information than previously considered. The findings bring new insight into the relationship between genotype and phenotype in Myxococcus xanthus.
Paschalidis recognized for developing intelligent systems that remain robust under uncertainty, enabling real-world use in medical data. His work showcases the power of convergent research in creating transformative impact on healthcare.
A new study reveals that AI systems mimic the structure of human judgment but with a more rigid, rule-based approach. The researchers found biases in AI judgments, especially across demographic traits, highlighting the need for awareness and understanding how these systems 'think'.
Researchers identified patient-reported symptoms associated with GLP-1s, including menstrual changes, fatigue, and temperature-related complaints, that may not be fully captured in clinical trials or drug labeling. Nearly 4% of Reddit users reported reproductive symptoms, and fatigue was the second most common complaint.
Researchers develop dynamics-driven models to identify disease transitions before symptoms appear, transforming real-time care and personalized treatment. AI systems analyze health data to detect
Researchers propose a temporally developmental continual learning framework inspired by human brain development, enabling cross-domain learning across perception, motor control, and interaction tasks. The approach achieves stable and strong continual learning performance while reducing network size and mitigating catastrophic forgetting.
A new study introduces AI-based control strategies that ensure local grids remain reliable and resilient. By utilizing Artificial Neural Networks, the system can predict and compensate for grid changes in real-time, outperforming traditional control methods.
Scientists at the University of Virginia Health System have developed a suite of AI-powered tools, called YuelDesign, YuelPocket and YuelBond, to transform how new drugs are created. These tools can design drug molecules tailored to fit their protein targets exactly, even accounting for protein flexibility.
Researchers from Tohoku University examined the role of materials databases in supporting modern artificial intelligence tools used in materials science. They found that database architecture can directly affect AI model performance and reliability. The study aims to improve database quality, connectivity, and develop new AI systems th...
UCLA investigators present new research on targeted drug delivery for colorectal cancer, COVID-19's impact on breast cancer outcomes, and AI in cancer diagnosis. These studies offer insights into overcoming drug resistance, enhancing immune responses, and improving outcomes for patients with difficult-to-treat cancers.
Several open-source AI models consistently produced summaries that were more comprehensive than physician-written versions, particularly in capturing molecular and genetic findings. The study found that Llama 3.1 and DeepSeek performed best among the six evaluated models.
Matei Zaharia's work on open-source systems like Apache Spark, Delta Lake, and MLflow has enabled large-scale machine learning, analytics, and AI at global scale. His innovations have made scalable computing accessible to researchers, nonprofits, and enterprises across every industry.
A study of 1,200 musicians in five countries reveals the 'streaming paradox': while artists are more visible than ever, meaningful income remains elusive. Musicians face growing pressure to create engaging content and manage their online presence.
A new survey by Ohio State University Wexner Medical Center finds that only 42% of Americans are open to using AI in their healthcare, down from 52% in a previous survey. The decline in trust is attributed to concerns about AI's accuracy and the potential for it to replace human judgment in medical decision-making.
National Academy of Engineering President Tsu-Jae Liu presents AI as a tool to expand engineers' problem-solving capacity and develop innovative solutions. The editorial calls for multidisciplinary education integrating AI while addressing its limitations and societal implications.
The article highlights the urgent need to modernize scientific record with AI-powered discoveries trapped in a predigital system. Top-tier research universities face $10-15 million annual subscription expenditures, while author-facing processing charges range from $5,000 to over $11,000 per article.
A team of researchers used AI to analyze 289,513 global publications and isolated 4,177 highly relevant studies on China's major basins. The study highlights crucial milestones in Chinese hydrology and aims to guide future research priorities and sustainable water resource strategies worldwide.
Gladstone Institutes investigator Ryan Corces receives $750,000 to investigate unknown genetic variants contributing to Alzheimer's disease. He aims to identify new drivers and therapeutic targets using artificial intelligence and CRISPR tools.
Researchers developed Sandook, a software-based system that tackles three major sources of performance-hampering variability simultaneously. The two-tier architecture optimizes task distribution for the overall pool while faster schedulers on each SSD react to urgent events.
A new study reveals teachers tend to provide regular support to students they've previously helped, rather than targeting all students equally. This finding highlights the need for AI-powered tools that can help teachers track their interactions and allocate time effectively.
The survey maps egocentric vision's major tasks, including subject understanding, object understanding, environment understanding, and hybrid understanding. It identifies bottlenecks such as limited datasets and dynamic first-person video, and highlights the need for stronger data, clearer task definitions, and deeper integration.
A new study showcases a self-supervised AI framework that can detect concrete cracks with strong accuracy and cross-dataset generalization, outperforming several widely used supervised deep learning models. The framework achieves this without relying heavily on large labeled datasets.
The University of Tennessee at Knoxville has received a $300,000 NSF grant to develop quantum computing-based tools for two-step uncertainty optimization problems. This will enable researchers and industry engineers to quickly determine the potential benefits of quantum computing in solving complex decisions.
A new study from CAMH reveals that AI models used to predict aggressive incidents in psychiatric care can amplify existing social and structural inequities. The model showed higher false positive rates for Black and Middle Eastern individuals, men, and those with unstable housing, potentially leading to more distress and loss of trust.
The UNU Hub for AI-Finance at East China Normal University's Shanghai AI-Finance School aims to leverage AI to boost sustainable economic development and financial inclusion. The Hub will prioritize innovation-driven research, global talent training, and international policy advocacy to address the Global South's financial challenges.
The UNU Hub will foster interdisciplinary collaboration, innovation, and policy engagement to address AI's ethical challenges. It aims to establish trusted frameworks ensuring AI serves as a force for equity, sustainability, and human well-being, aligned with UN Sustainable Development Goals.
The AUMOVIO-NTU Corporate Lab will focus on areas like AI, sustainability, novel materials, and connectivity to develop solutions for safer, smarter, and more sustainable transportation. The joint lab aims to accelerate innovations that can support Singapore's future mobility ecosystem.
A new study from Sultan Qaboos University demonstrates how AI can analyze Oman's Labour Law of 2023, revealing complex interdependencies between its articles. The research identifies influential 'hubs' within the law that may be triggered by changes, enabling policymakers to anticipate broader impact.
The Alliance for Clinical Trials in Oncology is enrolling adolescent and young adult cancer patients in various trials, including genetic services and treatment studies. These trials aim to address longstanding gaps in care and improve outcomes for AYAs with cancer.
University of Utah researchers develop framework for assessing varying levels of automation in psychotherapy, highlighting the importance of weighing risks and benefits. They explore how automation can support therapists without replacing them, using large language models to capture core components of treatment and provide feedback.
HSE researchers train a neural network, GSMFormer-PPI, to predict protein–protein interactions by integrating three types of data: sequence, structure, and surface properties. The model achieves 95.7% accuracy, outperforming popular graph-based models.
AI models exhibit emergent intelligence through specialization and cooperation, unlike physical systems where individual components reflect similar information, according to Bar-Ilan University research. This finding has implications for neuroscience and challenges traditional notions of intelligence in AI.
A new testing framework, Scalable Experimental Design for System-level Ethical Testing (SEED-SET), balances measurable outcomes and qualitative values like fairness. The system uses a large language model to capture stakeholder preferences and identifies scenarios where AI systems align with human values.
A new study provides a solutions-focused pathway to implementing the High Seas Treaty, highlighting the need for enhanced data resources and sharing. The researchers identify major scientific and technical developments that can help address challenges in biodiversity monitoring and connectivity between areas.
The article highlights the dangers of autonomous AI systems interacting within clinical environments, creating a 'digital ecosystem' that can operate beyond human oversight. The analysis warns of risks including the rapid propagation of errors, accelerated data leaks, and the spontaneous development of unintended hierarchies.
A new portable AI-powered scanning slit-light device makes low-cost eye disease screening more accessible, providing quantitative measurements and qualitative assessments of anterior-segment abnormalities. The lightweight AI model allows for on-device analysis, reducing operator dependence and improving portability.
The Association for Computing Machinery has published its inaugural issue of ACM AI Letters, a premier venue for rapid and timely AI research. The journal aims to bridge the gap between traditional conferences and journals, featuring short, peer-reviewed contributions that accelerate knowledge dissemination across academia and industry.
The São Paulo School of Advanced Science on Regulatory Governance explores emerging regulatory frontiers in Latin America. The program features speakers from leading international institutions and offers a range of activities, including flash talks, poster presentations, and paper development workshops.
Researchers propose a foundational framework to help multi-agent, connected systems decide what information they can trust before acting. The 'cy-trust' concept assigns a numerical trust value between 0 and 1 to data from other agents based on sensing, context, network behavior, and past experience.
Researchers at Tohoku University demonstrated that living biological neurons can be trained to perform a supervised temporal pattern learning task. The study integrates cultured neuronal networks into a machine learning framework, generating complex time-series signals comparable to those involved in motor control.
The adoption of AI-powered scribes was associated with modest decreases in total electronic health record time and documentation time. This is due to automation of routine tasks allowing clinicians more time for high-value patient care.
Current AI systems lack internal embodiment, a property that humans take for granted, which can lead to performance and behavior limitations. Researchers propose a dual-embodiment framework to guide future research in building safer and more aligned AI models.
A multi-site study found that AI scribes reduced electronic health record usage by 3% and clinical documentation time by 10%, with primary care physicians experiencing the greatest reductions. The technology also showed a slight increase in productivity, measured as 0.5 additional patient visits per week.
Researchers at KIT analyzed scientific papers using AI to identify emerging trends and relationships. The study suggests that large language models and machine learning can help reveal new avenues of research and opportunities for interdisciplinary cooperation.
A new AI chatbot helps patients access retinal detachment advice through personalized, real-time, clinically grounded conversations. The system outperformed leading large language models and includes accessibility features for people with low vision or limited English proficiency.
Research warns AI can compromise creativity, critical thinking with overly automated learning systems; 'learning vaults' suggested to safeguard adaptive knowledge. HR teams must ensure continued access to first-hand learning and human interaction.
A Kennesaw State University project is exploring AI-powered chatbots to help future teachers respond to how students think about math. The chatbots simulate classroom interactions, allowing student-teachers to practice responsive teaching and develop their skills in a more authentic way.
Binghui Wang, an Illinois Tech Assistant Professor of Computer Science, has been awarded the 2025 Distinguished Junior R&D Award by the IEEE Chicago Section for his groundbreaking research in AI security and trustworthy machine learning. The award recognizes Wang's impactful work on developing provably secure and trustworthy AI systems.
The NIH has renewed support for the USC-led AI4AD initiative, expanding its efforts to better classify Alzheimer's and related diseases, predict progression, and identify new treatment targets. The $12.6 million award will develop AI tools to uncover biological causes of Alzheimer's and improve disease diagnosis.
Researchers develop a smarter image-generation system that produces accurate, high-quality building designs. The framework uses real architectural datasets to generate detailed sketches and renderings, overcoming previous limitations.
Researchers have created a new AI model that can simulate molecules under extreme conditions, allowing for reliable discoveries in fields like drug development and sustainable chemistry. The model's stability opens up new opportunities for simulations in areas where long-term accuracy is essential.
Researchers evaluated the performance of the Phoenix Sepsis Score for predicting in-hospital mortality among pediatric ICU patients in China. They found that a modified version, PSS+, showed substantially improved discrimination without sacrificing clinical usability.