Articles tagged with Artificial Intelligence
MIT researchers developed a generative AI-driven approach for planning long-term visual tasks, surpassing existing techniques with a 70% success rate. The system combines vision-language models with formal planners, enabling robots to navigate complex environments and assemble multi-robot teams with high efficiency.
Researchers at MIT developed a new method that coaxes AI models to achieve better accuracy and clearer explanations in safety-critical applications. The approach extracts concepts the model has learned while training for a specific task and forces it to use those, producing better explanations than standard concept bottleneck models.
New research published in Weather assesses the energy consumption of AI models and finds that they consume considerable energy during training, but offset this by rapid forecasting ability. AI data-driven models are estimated to consume at least 21 times less energy than traditional models over one-year usage.
Jeonbuk National University researchers have developed DDINet, a lightweight and scalable model that can accurately predict drug-drug interactions for new, unseen drugs. This approach avoids overfitting to training data and is designed to handle binary and multi-classification tasks.
AI chatbots are homogenizing human expression and thought, reducing collective wisdom and adaptability, researchers argue. Incorporating real-world diversity into LLM training sets can help preserve human cognitive diversity and improve chatbots' reasoning abilities.
Researchers at Cornell University found that AI writing tools can shift users' attitudes on societal issues, even when they're aware of the bias. The study involved over 2,500 participants who used biased AI writing assistants and saw their views gravitate toward the AI's positions.
Emerging research across conceptual frameworks, biomarker science, digital phenotyping, and artificial intelligence synthesizes a translational pathway toward a more biologically grounded and clinically useful approach to psychiatric diagnosis. The current system falls short due to standardized clinical language and lack of biological ...
A study found that radiologists who reviewed imaging tests twice after AI flagged them as abnormal were 53% less likely to be held liable in court compared to those who reviewed once. This suggests that changes to radiologist workflow could reduce legal risk, but may also come with increased costs.
A novel AI-based framework, EDGCN, decodes dynamic variations in EEG patterns for better brain-computer interface technology. The model outperforms current state-of-the-art methods with high classification accuracies and decoding accuracy, enabling more seamless communication between the human brain and machines.
A Chinese Neurosurgical Journal study developed a radiomics-based machine learning model to identify high-risk patients with traumatic brain injury who require emergency decompressive surgery. The model accurately distinguished patients who later required secondary surgery, suggesting its potential to complement clinical judgment.
Researchers developed Zephyrus, an AI agent capable of analyzing and answering questions in natural language about weather and climate data. The agent can handle language-based queries, translating them into code and generating plain language answers.
The Global Physics Summit will feature over 12,000 individual presentations on new research in astrophysics, particle physics, and quantum information science. Registered journalists and public information officers will receive daily emails with information during the meeting.
The partnership aims to advance embodied intelligence and humanoid robotics, focusing on whole-body motion control, dexterous manipulation, and brain-eye-hand coordination. The collaboration will drive practical implementations in areas like intelligent robotics and brain-computer interfaces.
The University of Hong Kong's MeckUp Quest 2026 AI Robotics Challenge brought together 19 elite student teams to develop intelligent robots for post-typhoon disaster response. The competition demonstrated the potential of youth-driven innovation to empower emergency rescue efforts and safeguard social good.
Researchers developed a TwinGuard defense framework that uses real-time digital twin and reinforcement learning AI to detect and neutralize 5G cyber-attacks. The system detected and blocked attacks in under 100 milliseconds, including handover flooding and E2 subscription flooding attacks.
A major UK study found that local councils are progressing at varying speeds towards AI adoption, with some councils building robust data foundations and others struggling with legacy systems. The report highlights the importance of leadership ambition, governance discipline, and strategic clarity in determining AI readiness.
A study examines the occupational health and safety implications of AI tools, revealing psychosocial, ethical, and privacy risks. The rapid integration of AI into the workplace has often outpaced the adaptation of regulatory and ethical frameworks, highlighting the need for a people-centered approach.
Cognitive neuroscientists are integrating genetics and AI to study language development and disorders. AI-based models can predict language development in children, while genetics research links rhythm disorders and dyslexia. The brain's wiring connecting language regions also reveals that language is a system, not a single 'thing'.
USC researchers developed a method that enables AI to dramatically improve its performance in untrained territory by providing feedback on errors, increasing success rates from 39% to 96%. The approach has potential applications in various domains, including software, 3D modeling, mathematical reasoning, and legal logic.
Researchers have developed a robotic gripper called ChicGrasp, which can learn to handle chickens by imitating human movements. The system uses advanced imitation learning algorithm and camera perceptions to grasp a chicken carcass by the legs and lift it on a shackle conveyor.
A recent article highlights the need for human moral agency in nursing, as AI systems simulate empathy but lack sentience and accountability. Nurses must be active leaders in AI design and implementation to preserve public trust and ensure compassionate care.
Researchers developed magnetically controlled microrobots made from diatoms to target glioblastoma lesions with photodynamic therapy. The microrobots achieved a significant cytotoxic effect on primary glioblastoma cells and demonstrated good biocompatibility.
The policy brief outlines four courses of action to tackle AI implementation: bottom-up, problem-driven planning, transparency, accountability, and oversight. Without these measures, AI in the public sector will amplify current dysfunctions and feed distrust.
Researchers assessed whether people can distinguish between AI-generated and human speech, discovering that short training minimally improves this ability. However, neural responses became more distinct for human versus AI speech, suggesting training can help in the future.
Engineers at the University of Pennsylvania have developed LIBRIS, an automated microfluidic platform capable of generating lipid nanoparticle formulations at high speed and scale. This enables the creation of large, systematic datasets needed to train predictive AI models, accelerating the design of lipid nanoparticles for mRNA delivery.
Generative AI in medical imaging enables data synthesis, image enhancement, and modality translation to improve diagnosis and treatment outcomes. The review discusses challenges and future prospects, including convergence with large-scale foundation models and multimodal information for holistic understanding.
A study published in the European Heart Journal found that AI analysis of mammograms can predict the risk of serious heart disease in women. Women with mild calcification were 30% more likely to suffer serious cardiovascular disease, while those with severe calcification were two to three times more likely.
Scientists at the University of Sydney have developed an ultra-compact AI chip that harnesses the power of light to perform calculations, potentially lowering energy consumption and increasing speed. The prototype, built in-house, achieved 90-99% classification accuracy in image classification tasks.
Researchers created a novel approach for simultaneous ERG, PanCK, and H&E image generation from label-free tissue sections, enhancing vascular invasion assessment accuracy and efficiency. The virtual multiplexed immunostaining method overcomes traditional IHC limitations, such as section-to-section variability and tissue loss.
The Double-Helix Logic of Curriculum proposes a structural theory redefining education in AI age, emphasizing coexistence of universality and personalization. This framework situates curriculum within a broader shift towards Human Interdependence Paradigm, structuring conditions for differentiated strengths to accumulate and reinforce ...
The Ateneo Laboratory for Intelligent Visual Environments (ALIVE) is developing machine learning solutions with industry partners to improve public health, traffic systems, and more. By bridging the gap between messy reality and mathematical models, ALIVE is creating intelligent visual systems that can handle real-world conditions.
A machine-learning model developed by Weill Cornell Medicine predicts preeclampsia risk in late pregnancy with high accuracy. The model uses electronic health record data to continuously update risk predictions, potentially providing clinicians with timely clinical action.
Researchers have advanced humanoid robot control, motion planning, and perception using Valkyrie, a 10-year mission robot developed by NASA. The university will continue research with Talos, a 1.75-metre-tall robot, to study human-robot cooperation and adapt to changing environments.
Researchers at Northwestern University have developed AI-designed robots called 'legged metamachines' that can combine and recombine in the wild, recover from injury and transform into new shapes. The robots can adapt to the environment, survive catastrophic damage and even recover from being chopped in half or cut into pieces.
Artificial intelligence can complement human expertise to improve radiology workflows and patient care. The JACR Focus Issue on AI in Workflow Optimization explores the integration of AI technology to enhance efficiency and deliver better care.
A new study suggests that over 40% of UK adults are happy to use ChatGPT for mental health support due to long waiting times for traditional services. However, experts caution that AI should not replace human healthcare professionals and raise concerns about the potential impact on education and physical health.
The University of Ottawa Heart Institute, McGill University, and the University of Ottawa have launched ARCHIMEDES, a national health data platform providing Canadian researchers with secure access to diverse health data. The platform enables collaboration, supports advanced analyses, including AI algorithms, and prioritizes public trust.
Healthspan Horizons links real-world longitudinal data with deep discovery science to create uniquely powerful, long-term datasets revealing what drives human healthspan over time. The platform aims to translate discoveries into clearer guidance on what helps people stay resilient, supporting more years of energy, strength, and indepen...
Tohoku University researchers use large AI models to explore vast chemical spaces and predict catalytic performance before synthesis. This approach accelerates discovery, reducing trial and error methods.
A study by Dr Marigo Raftopoulos found that 71% of professionals think AI performs 'very' or 'extremely' well due to its ability to save time and improve processes, accuracy and efficiency. However, 84% are concerned about the risks associated with AI, including cyber security, job losses, and military uses.
Florida Atlantic University has received a $4.5 million grant from the US Air Force to establish a high-fidelity platform for autonomous decision-making and real-time sensor fusion research. The T-1A Jayhawk simulator will be used to study cognitive performance, situational awareness, stress and decision-making under pressure.
Scientists have identified five distinct species of antbirds, including two new species, by integrating AI, vocal analysis, and traditional museum work. The study reveals that populations separated by major Amazonian rivers have evolved into distinct species.
A new AI-powered tool, Merlin, has been developed to assess 3D abdominal CT scans, identifying anatomical features and predicting disease onset years in advance. The tool surpassed traditional automated tools in tasks such as diagnosing diseases and predicting patient outcomes, with accuracy rates of up to 81%.
The AI model can accurately identify disease-causing mutations in human genes and is capable of designing new genomes that are as long as the genomes of simple bacteria. Evo 2 has been trained on over 100,000 species across the entire tree of life and can process genetic sequences of up to 1 million nucleotides at once.
A European consortium is developing AI-powered therapy assistants to help doctors adhere to medical guidelines. The GUIDE-AI project aims to reduce the complexity of guideline navigation, particularly for chronic diseases such as heart failure and asthma.
A new study explores the market effects of unleashing generative AI on creative endeavors, finding that low-quality 'AI slop' harms consumers and professionals. High-quality AI, however, can enhance professional work while providing consumers with better content options.
A recent study found that AI programs are up to 97% accurate at detecting pictures of deepfake faces, but performed at chance levels when it comes to detecting deepfake videos. Humans correctly identified real and fake videos about two-thirds of the time, picking up on subtle inconsistencies in movement and facial expressions. The stud...
A study at George Mason University found that nursing students in an AI-enhanced health informatics course showed statistically significant learning gains compared to traditional students. The course used AI-generated videos and clinically contextualized assignments to improve student engagement and outcomes.
Research participants reading AI-generated summaries of historical events showed more liberal opinions than those reading Wikipedia summaries, indicating a subtle influence of AI biases on public understanding of the past. The study highlights the need for LLMs to be designed with bias-reducing features to provide unbiased information.
Researchers at Weill Cornell Medicine will develop an AI system to identify patients with early signs of treatment-resistant prostate cancer. The project aims to improve clinical trials by predicting treatment outcomes and avoiding ineffective treatments for patients with specific tumor subtypes.
A new study proposes a framework for connecting AI risk prediction systems with scheduling platforms to detect emerging risks and adjust project plans before delays occur. This approach aims to close the gap between early warning and actionable response, enabling more resilient project delivery.
Researchers from the University of Exeter argue that AI models are not as smart as claimed, citing a potential 'transferability crisis'. They found that generalisability is difficult to predict and benchmark metrics often overstate model performance. The study highlights the need for rapid testing in real-world applications.
Researchers developed a digital safeguard, My Music My Choice, to protect songs from generative AI cloning. The tool adds imperceptible changes to a song's waveform, making it difficult for AI models to replicate.
A new study by Cleveland Clinic and Dyania Health demonstrates the potential of large language models in accurately identifying patients who could benefit from clinical research. The AI system reviewed 1,476 patients and identified 46 as potential matches, resulting in a more diverse patient population.
The study reveals how AI is fundamentally shifting the paradigm for discovering high-performance batteries and efficient electrocatalysts. AI-driven generative models and Large Language Models are enabling scientists to predict exact chemical structures and navigate vast chemical spaces at unprecedented speeds.
New research warns that popular AI systems for predicting cancer biology may rely on visual shortcuts, raising concerns about their reliability for real-world patient care. Despite achieving high accuracy, these models often fail when conditions change or confounding factors are controlled.
Researchers at the University of Bath developed an AI tool that predicts building emissions from simple text descriptions, providing architects with real-time feedback on sustainability. The tool uses machine learning and natural language processing to estimate embodied carbon and suggests improvements to environmental conditions.
The journal argues that AI can process vast datasets to detect patterns and predict climate impacts, improving pollution tracking, climate modeling, and agricultural planning. AI-driven modeling can also help policymakers evaluate tradeoffs and anticipate unintended consequences, supporting more informed decision-making.
Recent advances in machine learning enable the non-targeted analysis of thousands of chemical features in a single environmental sample. Machine learning models can predict tandem mass spectra from known molecular structures and infer molecular formulas, significantly narrowing down candidate structures.
The study identifies three diverging global AI pathways, each shaped by distinct policy priorities, innovation models, and governance philosophies. The US prioritizes innovation speed, China emphasizes deployment scale, and the EU focuses on governance and societal safeguards.