Tech & AI

Researchers have built the first stable, bright near-infrared LEDs that work reliably at high power—a breakthrough that could accelerate development of night vision, medical imaging, and secure communications systems. The lead-free design solves a decade-old durability problem that has blocked commercial deployment.EN

Researchers have created an electrically charged growing substrate that accelerates plant growth in hydroponic systems, boosting seedling dry weight by half in just two weeks. The breakthrough could slash fertilizer use and upend indoor farming economics, potentially transforming agtech operations and food production costs.EN

Researchers have shown that infrared thermography can accurately measure how fast moisture leaves wood during drying—a finding that could help manufacturers optimize kiln operations and reduce energy costs. The method matches results from traditional physics-based calculations but offers continuous, real-time monitoring that existing approaches cannot provide.EN

Researchers have identified the most effective methods for autonomous vehicles to navigate underground mines using 2D laser scanners, with feature extraction techniques outperforming traditional approaches. The findings could accelerate deployment of autonomous mining equipment while reducing safety risks and operational costs in one of industry's most hazardous environments.EN

Researchers cracked a computational bottleneck in medical AI by showing that brain age predictions don't need expensive 3D scans—simple 2D projections work almost as well. For hospitals and researchers processing thousands of brain MRIs, the speedup could make large-scale studies feasible on standard hardware.EN

Scientists have developed a method to embed conductive patterns directly into paper using a 3D printer-like process, creating flexible supercapacitors that hold charge efficiently. The breakthrough could enable cheaper, thinner electronic devices for everything from wearables to energy storage—eliminating the need for multiple material layers that currently drive up production costs.EN

Researchers modeled lead emissions from glassworks across 20th-century Sweden, finding peak exposure in the 1970s-80s affected thousands of residents. The findings highlight how industrial air pollution persisted for decades before regulations caught up—a cautionary tale for today's emerging industrial regions and regulators weighing enforcement timelines.EN

Researchers discovered how to engineer titanium oxynitride films by adjusting oxygen and nitrogen ratios, controlling whether the material becomes transparent or light-absorbing. The findings could enable manufacturers to design custom optical coatings and components for next-generation displays and industrial applications without retooling production.EN

A new case study of a government ministry's AI chatbot found significant gaps between how officials expect citizens to respond and actual public sentiment. The findings suggest agencies deploying AI for citizen services may be underestimating trust concerns—a critical factor for adoption rates and service legitimacy.EN

Researchers have successfully modeled and tested supersonic combustion in scramjet engines—a critical breakthrough for the next generation of hypersonic aircraft. The work, which combines advanced simulations with real flight-scale experiments, could accelerate development of military and commercial hypersonic vehicles by providing engineers with reliable design tools they've lacked.EN

Researchers analyzing Turkey and Sweden's construction sectors have identified which factors most influence the industry's transition to sustainable practices—and crucially, which ones drive change in others. The findings could help policymakers and construction firms prioritize investments and regulations where they'll have the greatest ripple effect.EN

Researchers have developed practical techniques to repair and stabilize irregular salt caverns used for hydrogen and compressed air energy storage. The findings could accelerate deployment of large-scale energy storage systems critical for grid stability and renewable energy integration—a $10+ billion market growing rapidly as nations electrify.EN

Researchers found that five major disease-relevant enzymes pick their protein targets based on surrounding chemical context, not just known consensus patterns. The discovery could reshape how drug developers design therapies and predict off-target effects in precision medicine.EN

Researchers have created yeast-based microcapsules that cling to plant leaves and release pesticides only when needed, slashing waste by over 80%. The technology could significantly reduce chemical runoff while lowering farmers' input costs—a breakthrough that addresses both environmental and agricultural economics.EN

Scientists have demonstrated a new way to squeeze light into impossibly small spaces without the heat damage that plagues current methods, dramatically improving light-frequency conversion efficiency. The breakthrough uses a ceramic material called lithium niobate and could enable more compact, efficient photonic devices for telecommunications, sensing, and quantum computing applications.EN

Researchers have developed a forecasting system that predicts when parking lots will fill up by analyzing how drivers actually arrive and leave. The advance could help cities reduce traffic congestion and emissions by directing drivers to available spaces faster, while cutting the computational costs of existing prediction methods.EN

Engineers have struggled to predict when rubber components fail because existing fatigue tests vary wildly depending on loading conditions. A new mathematical approach promises consistent predictions regardless of stress patterns, potentially extending the lifespan of everything from car suspension mounts to industrial vibration isolators—and reducing costly field failures.EN

Researchers combined machine learning with ChatGPT-style AI to forecast safety risks at dams by comparing new structures to historical cases. The system achieved 78% accuracy on real-world Chinese dams, suggesting AI could help utilities and governments prevent costly failures and flooding without expensive on-site inspections.EN

Researchers developed an automated system that generates safer, more practical mitigation recommendations for dam operators by combining AI language models with a decision-tree algorithm. The approach improved recommendation quality by 94%, potentially reducing infrastructure failures and the operational burden on engineers managing critical water systems.EN

Researchers discovered that heating iron chalcogenide superconductors causes their prized quantum properties to vanish, even though the material's fundamental structure remains unchanged. The finding threatens the viability of these materials for next-generation quantum devices and suggests engineers must rethink how to engineer stable topological materials for real-world applications.EN

Researchers have discovered an unexpected connection between a new class of algebraic structures and the Jacobian Conjecture, one of mathematics' most famous open problems. The finding could reshape how mathematicians approach this decades-old puzzle, with potential implications for computational geometry and automated reasoning systems that rely on polynomial equations.EN

Researchers have cracked a major bottleneck in automated planning systems used for robotics, logistics, and autonomous systems. New compilation methods let AI planners handle safety rules and task dependencies on large, complex problems 100+ times faster than before—opening the door to practical deployment in manufacturing, supply chains, and other data-heavy industries.EN

Researchers have demonstrated that physically large antenna arrays can beam power to IoT devices reliably over distance while staying within regulatory limits. The breakthrough means battery-free connected devices could soon become practical for real-world deployments, addressing a major cost and maintenance barrier for enterprises rolling out IoT at scale.EN

Researchers have identified how atoms migrate along grain boundaries in 304L stainless steel when subjected to prolonged stress, a process that weakens the material over time. The finding could help engineers design longer-lasting components for power plants, pipelines, and aerospace systems—where unexpected failure carries enormous costs.EN

Scientists have created the first complete framework for understanding how quantum measurements work when computers share information over networks. The findings could accelerate development of practical quantum networks and help companies build more efficient quantum systems that rely on distributed computing and real-time communication.EN