AI narrows the productivity gap
Researchers found that an AI app improved efficiency for low-skilled taxi drivers by 7%, while high-skilled drivers saw little benefit. The study suggests AI can help close skills gaps in various professions.
Artificial Intelligence
Articles tagged with Artificial Intelligence
Instant AI-assisted test for viral infection
A non-DNA based test using a nanopore device and AI model can identify live viruses in patients within minutes. The system showed 100% sensitivity and 71.4% specificity for detecting herpesvirus, with improved accuracy when conditions were adjusted.
Tradition meets AI as Leicester scientists help tackle Amazonian biodiversity crisis
Scientists from University of Leicester collaborate with traditional Amazonian communities to monitor biodiversity using artificial intelligence, enabling them to maintain control over their resources and territories. The project aims to reshape conservation approaches in the world's largest tropical rainforest.
KAIST introduces ‘Virtual Teaching Assistant’ that can answer even in the middle of the night – successful first deployment in classroom
The virtual teaching assistant (VTA) provides personalized feedback to individual students even in large-scale classes. The system, which automatically vectorizes a large volume of course materials and uses them as the basis for answering students' questions, has been shown to significantly reduce the burden on TAs.
Robots made of linked particle chains
Researchers at Harvard developed link-bots, centimeter-scale robots composed of V-shaped chains with notched links, capable of coordinated movements and emergent collective behavior. The team demonstrated link-bots' ability to move forward, stop, change direction, squeeze through gaps, and cooperate on tasks.
Modeling electric response of materials, a million atoms at a time
Researchers developed a machine learning framework that can predict how materials respond to electric fields up to a million atoms, accelerating simulations beyond quantum mechanical methods. This allows for accurate, large-scale simulations of material responses to various external stimuli.
AI identifies key gene sets that cause complex diseases
Researchers developed a new computational tool using generative AI to identify key gene combinations underlying complex illnesses. The method amplifies limited gene expression data, enabling researchers to resolve patterns of gene activity that cause complex traits.
New book: Machine Learning in Quantum Sciences
The book, co-authored by 29 contributors from over ten countries, offers an introduction to machine learning and deep neural networks for complex quantum problems. It serves as a timely guide for PhD students and researchers looking to apply modern machine learning methods to quantum physics and chemistry.
Global bottom-up initiative takes off to map 80% of chronic disease: All health stakeholders herald a new era in diagnosis, prevention & treatment
The Human Exposome Moonshot initiative aims to map the physical, chemical, biological and psychosocial exposures driving 80% of chronic diseases. The exposome project integrates advanced technologies to create a comprehensive understanding of environmental influences on health.
JMIR Human Factors invites submission on human factors in health care
The journal JMIR Human Factors is inviting submissions for a new theme issue focusing on human factors in health care education, management, and knowledge translation. The issue aims to explore current and emergent educational and training aspects of human factors, including digital competencies for healthcare professionals.
Can AI accurately detect live oysters?
A new study published in Frontiers in Robotics and AI explores the use of artificial intelligence to detect live oysters. While the AI model ODYSSEE shows promise, it lags behind humans in accuracy, correctly identifying live oysters at a rate of 63% compared to 74% for expert annotators.
ELSI’s challenges in the era of advanced AI
The ELSI University Summit in Japan brought together experts to discuss challenges and prospects of responsible AI development. Key findings include the need for ethics-driven approaches, collaboration between academia and industry, and the importance of setting evaluation indicators for Responsible AI.
Urgent policy actions needed to address real AI threats, scientist reveals
Professor Shalom Lappin argues that tech companies' dominance in AI development poses a threat to public benefit and environment. He calls for comprehensive international regulation, intellectual property reform, and preparation for workforce disruption to address these challenges.
Scientific community urges greater action against the silent rise of liver diseases
Chronic liver diseases like MASLD and MASH affect 33% and 5% of adults worldwide, respectively. Experts propose doubling MASH diagnosis rate by 2027 to improve outcomes and reduce healthcare burdens. A paradigm shift towards preventive hepatology is key to addressing this growing public health threat.
Giving doctors an AI-powered head start on skin cancer
PanDerm, a new AI model, analyzes multiple types of images to improve skin cancer diagnosis accuracy by 11% when used by doctors. The tool also enhances diagnostic accuracy for non-dermatologist healthcare professionals by 16.5%.
AI agents for oncology: Dresden research team develops system to support clinical decisions
The Dresden research team developed an autonomous AI agent for precision medicine that can process multimodal data and have reasoning capabilities. The AI agent reached correct clinical conclusions in 91% of cases and accurately cited relevant oncology guidelines in over 75% of its responses.
New system enables robots to solve manipulation problems in seconds
Researchers developed an algorithm that lets a robot think ahead and consider thousands of potential motion plans simultaneously, solving multistep manipulation problems in a matter of seconds. The new method enables robots to rapidly determine how to manipulate and pack items without damaging them, even in narrow spaces.
Using swarm intelligence to improve treatment of acute stroke
A new computer model based on artificial intelligence is being developed to help doctors treat stroke patients. The model will use data from the German Stroke Registry and local brain images to predict long-term outcomes and potential complications, enabling doctors to make informed decisions about therapy.
USP and FAPESP bring innovative technologies to international startup fair in France
Researchers from USP and FAPESP will present their work on various topics, including health, agriculture, and artificial intelligence. The event aims to promote strategic connections and international visibility for startups and researchers.
Learning about autism from those who live on the spectrum
A study analyzing over 700,000 Reddit posts on autism reveals recurring topics such as music and social life, showcasing a supportive community that values differences. The research offers a new approach to understanding autism, shifting focus from disability to neurodiversity.
How and for whom using generative AI affects creativity: a field experiment
A field experiment by Tulane University researchers found that generative AI tools enhance employees' creativity when used by those who can manage their own thinking. Employees using ChatGPT performed better in generating novel and useful ideas.
Tech sector emissions, energy use grow with rise of AI
The ITU-WBA Greening Digital Companies report shows that tech sector carbon emissions continued to rise in recent years due to rapid advances in AI and data infrastructure. Despite this, the report highlights encouraging progress as more companies set emissions targets, source renewable energy, and align with science-based frameworks.
Cancer diagnosis on your laptop? New artificial intelligence model makes it possible!
A revolutionary AI model has been developed to diagnose lung cancer without relying on costly GPU servers or massive datasets. The ultra-lightweight model achieved a discrimination performance corresponding to an AUC value of 0.92, outperforming state-of-the-art large-scale AI systems.
New AI transforms radiology with speed, accuracy never seen before
A new AI system has been developed that improves radiology productivity and accuracy, offering a breakthrough solution to the global radiologist shortage. The tool boosts efficiency by up to 40% without compromising accuracy and flags life-threatening conditions in real-time.
An algorithm reveals how our brain is motivated
A team from UNIGE, Harvard, and McGill has discovered that the ventral tegmental area (VTA) encodes not only anticipated reward but also precise moment of expectation. This finding highlights the value of combining AI with neuroscience to reveal brain mechanisms.
Our online world relies on encryption. What happens if it fails?
Researchers at Boston University are developing a groundbreaking method for securing sensitive data in the face of emerging quantum computing threats. Their approach, called Encrypted Operator Computing (EOC), merges physics, computer science, and mathematics to enable scalable methods for computing directly on encrypted data.
Many possible futures: How dopamine in the brain might inform AI that adapts quickly to change
Researchers found that brain's dopamine neurons encode a map of possible future rewards across time and magnitude, guiding adaptive behavior in uncertain environments. This biological insight aligns with recent advances in AI, particularly distributional RL algorithms, which learn from reward distributions rather than averages.
New research finds specific learning strategies can enhance AI model effectiveness in hospitals
Researchers used proactive and transfer learning strategies to mitigate data shifts in AI models for hospital applications. They found that models trained on one hospital type performed better than those trained on all hospitals using transfer learning.
‘Eye’ on health: AI detects dizziness and balance disorders remotely
Researchers developed an AI model that uses real-time data to diagnose nystagmus, a condition linked to vestibular or neurological disorders. The system allows patients to record their eye movements using a smartphone and receive remote diagnostic analysis.
GPT-generated educational materials for urological cancer patients, translated by AI into five languages, are rated by doctors as easier to read than human-authored versions while being just as clear, accurate and complete
Doctors rate GPT-generated educational materials as clear, accurate and complete as human-authored versions, but easier to read. The AI-generated materials were translated into five languages for a randomized assessment.
Unlocking the timecode of the Dead Sea Scrolls
A new AI-powered model, Enoch, uses radiocarbon dating and machine learning to accurately date individual Dead Sea Scrolls, pushing the known timeline of their creation. The study reveals that many scrolls are older than previously thought, changing our understanding of ancient Jewish script styles and literary origins.
Texas A&M launches innovative AI And Business Minor
The new minor will equip students with cutting-edge AI and machine learning skills essential for modern business leadership. It combines hands-on experience with practical applications of artificial intelligence in business settings.
Why AI can’t understand a flower the way humans do
A recent study by Ohio State University researchers found that AI lacks the human understanding of flowers due to its inability to experience sensory and motor aspects. This limitation affects not only flower but also other complex human concepts.
Researchers uncover hidden airport hotspots in global wildlife trafficking using AI
A study published in Communications Earth & Environment reveals how AI can help combat global wildlife trafficking by identifying key airport nodes. Researchers analyzed nearly 2,000 airports and predicted 307 potentially involved airports, including two US airports that had not been previously flagged. The model also identified signif...
Artificial intelligence for the most precise measurements possible
Researchers used AI to approach the fundamental limit of precision in optical methods, calculated using Fisher information. The team's algorithm achieved impressive results, only minimally worse than the theoretically achievable maximum, demonstrating its effectiveness.
University of Washington grad receives ACM Doctoral Dissertation Award for developing machine learning algorithms to improve mental health
Ashish Sharma's dissertation developed fundamental advances in natural language processing to positively impact mental health. His AI-supported tools have been used by over 160,000 people, with more than fifty percent of users reporting a household income of less than $40,000 per year.
AI analysis of ancient handwriting provides new age estimates for Dead Sea Scrolls
A new AI program, Enoch, has been developed to analyze ancient handwriting styles and estimate the age of undated manuscripts. The program, trained on radiocarbon dating data, was tested on 135 Dead Sea Scrolls and found realistic estimates for approximately 79% of the samples.
APA calls for guardrails, education, to protect adolescent AI users
The American Psychological Association recommends prioritizing features that protect adolescents from AI's negative effects, including exploitation and manipulation. The report suggests age-appropriate defaults in privacy settings and encourages the use of AI to promote healthy development.
‘AI scientist’ suggests combinations of widely available non-cancer drugs can kill cancer cells
A team of researchers used AI and human collaboration to identify six promising drug pairs against breast cancer cells. The combinations, including simvastatin and disulfiram, showed potential for further research in therapeutic repurposing.
Brains vs. bytes: Study compares diagnoses made by AI and clinicians
A University of Maine study evaluated over 7,000 anonymized medical queries and found that AI models performed well on factual and procedural queries but struggled with 'why' and 'how' questions. Human clinicians consistently outperformed AI in terms of emotional engagement and empathetic nuance.
Using AI to construct on the level
Researchers developed FLAT, a digital tool assessing laser scans to locate uneven areas on concrete slabs. The AI-assisted algorithm reduces foundation measurement time by over 90%, enabling faster building construction with minimal human involvement.
Listening to life: Speech technology transforms clinical research
Dr. Deanna Kaplan's innovative voice-capture app, Fabla, captures unstructured voice narratives to study how clinical interventions influence daily life. The platform has found applications across diverse health domains, including veteran experiences and healthcare provider burnout.
Attachment theory: A new lens for understanding human-AI relationships
Attachment theory is used to understand human-AI relationships, revealing anxiety and avoidance tendencies. The study suggests AI chatbots can be tailored to meet users' emotional needs, promoting psychological well-being.
AI chatbots aren’t experts on psych med reactions — yet
Researchers found that AI chatbots have difficulty detecting adverse drug reactions and providing personalized advice. The study suggests that improving AI for mental health concerns could be life-changing for communities with limited access to healthcare.
Patients say “Yes..ish” to the use of AI in dentistry
A multinational study involving Aarhus University sheds light on patients' attitudes toward AI in dental diagnostics. Patients view AI as a useful diagnostic tool but express concerns over data privacy and cost implications. The overwhelming majority insists on professional human oversight, echoing the views of dental professionals.
Self-powered artificial synapse mimics human color vision
Researchers at Tokyo University of Science developed a self-powered artificial synapse capable of distinguishing colors with remarkable precision. The device generates electricity via solar energy conversion, making it suitable for edge computing applications.
A novel, multimodal approach to automated speaking skill assessment
Researchers developed a novel approach to automated spoken English assessment integrating acoustic, turn-taking, linguistic, and visual components. The multioutput learning framework achieved an accuracy of approximately 83% in predicting SEE scores.
New hydrogenation reaction mechanism for superhydride revealed by machine learning
Researchers successfully reproduced high-pressure synthesis reaction of superhydrides using a machine learning model, revealing a unique reaction pathway involving surface melting, hydrogen absorption, and solidification. This breakthrough deepens understanding of high-pressure physico-chemical processes and holds promise for easier de...
Chatbot system simulates group therapy to manage premenstrual syndrome
A research team has designed a chatbot system to simulate group therapy for managing premenstrual syndrome (PMS), a common disorder among women. The system provides simultaneous support for understanding, coping, and connections with peers, improving engagement and language convergence with participants.
Transforming immunotherapy design
Assistant professor of electrical and computer engineering Natasa Miskov-Zivanov is receiving a $581,503 NSF CAREER Award for her project that leverages AI to design more effective lymphocytes for cancer immunotherapies. The system aims to accelerate the process of designing new therapeutic cell designs.
Traditional diagnostic decision support systems outperform generative AI for diagnosing disease
A new study compares traditional diagnostic decision support systems (DDSSs) to generative AI and finds that DDSSs perform somewhat better. The researchers envision pairing these two approaches to improve diagnostic capabilities and patient outcomes.
The future of AI regulation: why leashes are better than guardrails
Experts Cary Coglianese and Colton Crum argue that management-based regulation, or using
AI tool enables automated evaluation of facial palsy, reports Plastic and Reconstructive Surgery®
A refined artificial intelligence (AI) tool has shown promise for objective evaluation of patients with facial palsy. The 'fine-tuned' model demonstrated substantially lower error rates and improved keypoint detection in every area of the face, including areas of asymmetry.
Can AI make critical communications chips easier to design?
A multi-university team is using AI to design radio frequency integrated circuits (RFICs), reducing development time and cost. The project aims to lower the barrier to entry for researchers and companies, making RFICs more accessible to solve problems and advance technological innovation.
New AI tool reveals single-cell structure of chromosomes — in 3D
A new AI tool developed by University of Missouri researchers can predict the 3D shape of chromosomes inside individual cells, providing a new view of how genes work. The tool helps identify unique differences in chromosome folding between cells, which controls gene activity and can lead to diseases like cancer.
The African Engineering and Technology Network signs ninth university partner
The African Engineering and Technology Network has signed its ninth university partner, Universidade Agostinho Neto. The network, led by Carnegie Mellon University Africa, aims to promote pan-African digital growth through research, education, and entrepreneurship.
Hitting the right notes to play music by ear
The study identified four simple ways music learning technology can aid prospective musicians, including helping people improve recall while listening and replaying notes indefinitely. Researchers found that existing digital learning tools were not being used effectively by students despite their availability.
AI aversion in social interactions
An experimental study found that humans are less likely to exhibit fairness, trust, and cooperation when interacting with a large language model (LLM) compared to another human. Participants in online games played with either a human or an LLM acting on behalf of another person showed reduced social behavior.
Machine learning simplifies industrial laser processes
Researchers from Empa developed machine learning algorithms to optimize laser-based manufacturing techniques, reducing preliminary experiments by two-thirds. They also implemented real-time optimization using field-programmable gate arrays (FPGAs) for improved welding processes.
Can AI analogize?
A recent study evaluates GPT-4's performance in solving letter-string analogies, a task requiring basic counting and relational understanding. The model excels at this task, demonstrating emergent analogical reasoning capabilities that may be supported by structured operations and relational representations.