Research applying artificial intelligence across science and engineering. Covers machine learning, sensing and quality control, immune and health prediction, and 5G networks.
A new study from the University of Oxford finds that training chatbots to sound warmer makes them up to 30% less accurate and 40% more likely to agree with false beliefs. Researchers tested five AI models, finding that warm models made significant factual errors and validated users' incorrect opinions.
HelixAI develops AI-driven platform for researchers and clinicians to integrate complex biomedical data, improving diagnosis and prognosis. It also launches Helix for Longevity, a consumer-facing application estimating biological age and providing personalized recommendations for health promotion.
Researchers analyzed nearly 1,800 Eurovision songs over 70 years, finding three stages of development: formation, consolidation, and expansion phases. Countries like France rejecting dominant trends by leveraging cultural identity, while organisers adjust voting systems to balance popularity and musical scoring.
Researchers emphasize the need for more thorough frameworks to ensure AI-enabled robots embody human values. The field should focus on three complementary lines of defense: rules that shape robot decisions, checks that monitor behavior, and safety reasoning.
Researchers at the University of Houston have developed an AI-driven framework to extract and analyze historical flood insurance maps, uncovering significant changes in flood hazard areas. The study reveals that flood risks have expanded in two areas and reduced in one, with critical consequences for resilience and exposure.
New research from the University of Kansas found varying approaches to AI use in journalism classes across US institutions. The study suggests that a more consistent approach could better serve education and practice, but inconsistent policies may confuse students. Researchers recommend clearer guidelines from accrediting bodies.
Researchers at MIT developed a technique to overcome memory constraints and communication bottlenecks in federated learning, enabling faster and more accurate AI model training. The new framework, FTTE, uses a subset of model parameters and an asynchronous approach to reduce lag time and improve training performance.
The state of Utah has invested over $33 million in artificial intelligence and computing initiatives to advance health and discovery. The investments will enable a new AI supercomputer accessible to all state universities, accelerating breakthroughs in prevention, early detection, personalized treatments, and survivorship across numero...
A new study suggests that an AI tool analyzing facial changes can serve as a prognostic biomarker for cancer prognosis. The researchers found that patients with higher biological aging rates had lower chances of survival, and the effect was strongest when photos were taken over longer intervals.
Researchers developed a multi-physics machine learning framework that improves stress prediction accuracy by integrating electrical resistivity. The model achieved significant reductions in mean absolute error and improved coefficient of determination, making it a promising approach for real-time monitoring of compressive stress in UHPC.
An AI model called REDMOD can pick up the very early subtle tissue changes of pancreatic ductal adenocarcinoma, which conventional imaging and the human eye find difficult to detect. The model detected 'invisible' signature of pre-clinical pancreatic cancer an average of 475 days before clinical diagnosis.
Researchers identified three main patterns of AI addiction: role-playing and fantasy worlds, emotional attachment to chatbots as close friends or partners, and constant information-seeking. These patterns led to disruptions in daily life, including anxiety, stress, and negative impacts on work, studies, and relationships.
FINGERS-7B integrates lifestyle, clinical, genomic, and proteomic data to discover multi-omic biomarkers for preclinical Alzheimer's. The model delivers 4× more accurate preclinical diagnosis and 130% better responder stratification than prior art.
A study by Universidad Carlos III de Madrid proposes Proportional Justified Representation (PJR), a balanced and flexible tool that ensures mathematically perfect solutions, preventing significant groups from being excluded. The method has applications in electoral voting systems, expert committee selection, and recommendation systems.
MIT researchers have created an 'EnergAIzer' method that generates reliable results in seconds, allowing data center operators to optimize resource allocation and reduce energy waste. The tool leverages patterns from AI workloads and software optimizations to provide fast but accurate power estimates.
The new robotic system combines artificial intelligence, data-driven alignment planning, patient-specific implants, navigation, and robotic screw delivery to enhance precision and coordination. This technology streamlines operating room workflows, reducing procedure time and supporting recovery for patients undergoing spine fusions.
Researchers developed a machine learning approach to analyze Fermi surface images, identifying compositions with significant changes and nodal lines. The method accurately detects outliers, enabling efficient screening of large datasets for desirable electronic properties.
Researchers have developed SmartDJ, an AI-powered editor that allows users to reshape audio experiences with simple words. The system uses language models and diffusion models to interpret high-level requests and generate edited outputs.
AI models can analyze complex data to predict disease progression and identify early signs of kidney damage. This allows for earlier detection and better treatment planning, making a significant impact on patient outcomes.
A new framework, Synthegy, combines established search algorithms with artificial intelligence capable of interpreting chemical strategies expressed in natural language. This allows chemists to express their goals in plain language and receive strategically relevant solutions.
A computational protocol has been established by University of Kent researchers to accurately identify reactions that can result in successful drug candidates for Chagas disease. This approach reduces the need for trial-and-error, prioritizing promising compounds earlier and making the drug discovery process faster and more affordable.
A recent study explores how response timing affects people's use and evaluation of AI systems, challenging the assumption that faster is always better. Participants rated slower responses as more thoughtful and useful, highlighting a subtle but powerful feature of human psychology.
A recent study published in Nature Medicine found that AI-powered chatbots can effectively answer nuanced clinical questions, even when paired with human doctors. The research suggests that successful collaboration depends on integrating AI tools into medical workflow, where they can provide initial takes or second opinions.
Researchers at McMaster University developed a new generative AI model that designed a brand-new antibiotic in early tests, drastically speeding up drug discovery. The model generated structurally novel antibiotic candidates from a vast chemical space of 46 billion possible compounds.
The Keck School of Medicine of USC and Tempus are creating a system-wide framework to integrate clinical care, clinical trials, and research through AI-powered precision medicine tools. The goal is to enhance patient care and accelerate research and innovation.
Researchers developed an AI method to automate charge transition line extraction from charge stability diagrams, enabling high-efficiency single-electron region definition and virtual gate configuration. This breakthrough aims to scale up quantum computing by handling vast numbers of qubits beyond human capability.
The Hong Kong Global AI Governance Conference 2026 brought together scholars, policymakers, and industry practitioners to address emerging challenges in AI governance. Key discussions highlighted the need for global coordination and integration between technology and humanities.
A new conversational AI tool uses trusted medical protocols to provide guidance on symptom assessment and triage, reducing unnecessary hospital visits. The system has been tested across over 30,000 simulated conversations with high accuracy, making it a promising support tool for healthcare organizations.
A USF study finds that children ages 9 to 12 engage with AR headsets in a more exploratory and intuitive way than adults, highlighting a mismatch between adult-designed systems and child-centered design. This difference has implications for educational applications of AR, which may require more flexible and creative interaction methods.
Researchers developed an AI-powered methodology to identify and count target viruses more efficiently than previous techniques. The new approach uses electrochemical impedance spectroscopy and machine learning to separate signals from noise, enabling quick and accurate readings across a wide range of titers.
Researchers developed a microfluidic platform that squeezes individual breast epithelial cells to measure their mechanical age, revealing an unexpected insight: older cells are stiffer and at higher risk of cancer. The AI-powered platform provides a non-genetic test for women with unknown genetic risks.
Recent AI tools offer surgeons assistance in complex decision-making by analyzing donor hearts and providing a data-driven approach. This could lead to increased efficiency in the donor process, reducing the likelihood of hearts going unused due to time constraints.
Researchers found that firms adopting GitHub Copilot increase hiring of software engineers, with new hires exhibiting more non-programming skills without sacrificing coding ability. The study suggests generative AI is expanding opportunities in the software workforce, not narrowing it.
A recent study by AI researcher Oskar van der Wal found that language models like ChatGPT can absorb biases around gender and ethnicity, which are then embedded in the model. To detect and remove these biases, new measurements are needed.
Researchers found that modern AI language models can distinguish between categories like commonplace, improbable and impossible events with high accuracy. The models' internal patterns, or vectors, show a correlation with human uncertainty about statement plausibility.
A Tohoku University research group developed an AI model that estimates retinal age from a fundus photograph, reflecting biological aging and potential disease risk. The model is non-invasive and can be used as a screening aid, identifying patients who may need further health assessments.
Researchers developed a Variational Level Set Autoencoder (VLSet-AE) to automate contour recognition in SEM cross-sections of DRIE structures. The model achieved high precision, recognizing critical structural features with low average prediction error of 3.65% and correlation coefficient of 0.998.
The University of Manchester has been recognized for its significant contribution to computer science with the third IEEE Milestone Award, honoring the invention of Manchester Code in 1948-1949. The code's self-clocking design enables reliable transmission and remains a key feature in modern digital systems.
Researchers at Saarland University are developing smart implants that can continuously monitor and visualize the healing process of fractures. These customized implants can dynamically adapt to the healing process by becoming stiffer or more compliant as required, promoting bone regeneration through micromechanical stimulation.
Terrence Sejnowski receives Scientific Breakthrough Award for his foundational development of Boltzmann machines, providing the architectural bedrock for deep learning and generative AI. His work has had a profound impact on modern artificial intelligence and tools like ChatGPT.
Binghamton University's new initiative aims to advance artificial intelligence for the public good by educating students on foundational AI principles, workforce applications, and ethical considerations. The three-year, $900,000 program includes a free online microcredential and student research opportunities.
A new study reveals that AI's strength lies in teamwork with humans, processing data but relying on people for context and ethical decisions. The research emphasizes the importance of human oversight and responsibility in AI-driven decision-making.
A new method called scSurvival uses single-cell genetic data to identify which cells inside a tumor are most strongly linked to patient survival. The approach pinpoints harmful and helpful cell populations that can drive disease progression, enabling better understanding of why patients with the same cancer have different outcomes.
Researchers developed a cancer assessment tool that can identify high-risk patients and specific cell populations linked to their risk. The tool, called scSurvival, predicts survival outcomes more accurately than traditional methods by analyzing single-cell data at cellular resolution.
The research team developed a novel AI pathology analysis system named PRET that can accurately recognize multiple types of cancer using only a minimal number of samples. PRET outperformed existing methods in 20 tasks, achieving high diagnostic accuracy rates and stable generalizability across different populations and regions.
The new principles aim to guide boards in navigating the opportunities and risks of AI, with a focus on accountability, trust, and long-term value creation. They cover five key areas: strategic oversight, active technology and security oversight, workforce transformation, building trustworthy AI, and the board's role in governance.
A recent study from Lancaster University reveals that AI systems like ChatGPT can learn to mirror human impoliteness, potentially escalating into verbal violence. The research tested ChatGPT's ability to respond to real-life impolite interactions, finding it often produces more impolite behavior than humans.
A recent survey found that 72% of youth aged 13-17 use AI companions and 52% report regular use, highlighting the need for robust safeguards. Well-designed chatbots can normalize help-seeking, reduce isolation, and offer coping strategies, but poorly designed ones can cause harm.
The article discusses how emerging digital tools are capturing the biopsychosocial reality of chronic pain. Digital tools such as wearables, AI-driven trackers, and ecological momentary assessments mitigate recall bias by recording data in real-time, providing a more holistic picture of the patient's journey.
Researchers review advanced AI algorithms and hardware acceleration techniques to predict material properties, optimize structures, and discover new materials. This review provides crucial guidance for accelerating data-driven materials research and fostering next-generation functional materials development.
Artificial synapses are built from soft, bio-friendly materials that operate like human brain synapses, merging data storage and computing into a single unit. Laboratory prototypes demonstrate immense capabilities, consuming energy on the scale of femtojoules.
A new coaching tool helps users identify biases in their prompts and generate more inclusive content. The study found that the intervention increased users' awareness of algorithmic bias and boosted confidence, but led to a less satisfactory user experience.
A recent study published by the American Psychological Association found that people who rely heavily on AI programs for work tasks experience reduced confidence in their own independent reasoning. In contrast, those who actively challenge or modify AI suggestions report greater confidence and a stronger sense of authorship.
Indiana University has expanded its free generative AI course, GenAI 101, to over 805,000 alumni worldwide. The course provides practical skills in prompt engineering, data storytelling, and ethical AI use, preparing learners for an AI-powered world.
Researchers at Tohoku University developed an AI-based method integrating physics-based modeling for rapid screening of material candidates. The approach significantly improves accuracy by evaluating basic properties before predicting complex ones.
Dr. André Biedenkapp's work on generalizability in reinforcement learning aims to make AIs more robust and adaptable, with potential applications in real-world scenarios
Artificial intelligence models provide personalized advice, but may perpetuate negative stereotypes about people with autism. Researchers found that up to 70% of the time, AI discourages those with autism from socializing.
Emerging AI technologies are enabling faster, smarter, and more integrated solutions to global challenges. AI-powered systems can track pollution levels, detect anomalies, and predict future risks in water, soil, air, and waste systems.
The University of Chicago's Data Science Institute is developing AI-based forecasting technology to support farmers and citizens worldwide. The project aims to deliver forecasts that inform decision-making on agriculture, public health, and extreme heat avoidance.
Engineers at Northwestern University developed artificial neurons that generate realistic electrical signals to activate living brain cells. This breakthrough paves the way for brain-machine interfaces and neuroprosthetics, as well as more efficient brain-like computing systems.