Tech & AI

Researchers testing solar and heat pump combinations in Sweden found that achieving positive energy district status—where buildings produce as much energy as they consume—is not feasible with current technology. The finding challenges urban planners and developers betting on these systems to meet climate targets without grid upgrades.EN

A new study of artisanal miners in Ghana documents systemic sexual harassment, coercion, and abuse that keeps women trapped in exploitative work. The findings challenge development programs that overlook workplace safety—a critical oversight for companies and governments targeting female economic empowerment in resource-rich regions.EN

Researchers have optimized titanium-aluminum-nitride coatings that achieve record hardness levels, potentially extending the lifespan of cutting tools, drill bits, and engine components. The breakthrough shows how adjusting aluminum content and manufacturing conditions can fine-tune coating performance—offering manufacturers a path to reduce tool replacement costs and production downtime.EN

Researchers have developed an artificial intelligence system that predicts short-term power demand more accurately than existing methods, combining neural networks with attention mechanisms and quantile regression. The advance matters because utilities lose billions annually to forecast errors that disrupt grid operations and inflate energy costs.EN

Researchers have engineered tiny particles that deliver cancer-fighting drugs directly to tumor cells while protecting healthy tissue, a breakthrough that could improve treatment options for aggressive breast cancers. The approach cuts drug toxicity while increasing its potency, potentially opening a new market for precision oncology treatments.EN

Researchers found that heating biomethanation reactors to 70°C dramatically improves methane yields from carbon dioxide and hydrogen—a 24.6% productivity gain. This discovery could accelerate commercial development of renewable methane technology, offering energy companies a practical route to store renewable energy and reduce emissions simultaneously.EN

Researchers have successfully used atmospheric plasma spraying—a rapid, high-temperature coating method—to build the key layers of solid-state batteries. If the approach proves commercially viable, it could bypass expensive manufacturing bottlenecks and accelerate adoption of next-generation batteries for electric vehicles and energy storage.EN

Researchers developed a machine learning system that predicts how manufacturing changes will affect production before costly real-world tests. The approach cuts design time for complex factories and could help manufacturers respond faster to demand swings and quality issues—a significant advantage in competitive sectors like marine engine production.EN

Researchers have designed a microdevice that separates blood cells using sound waves, potentially enabling faster, cheaper diagnostics at the point of care. The simulation shows the technique can isolate white cells, red cells, and platelets simultaneously—a capability that could reduce testing time and equipment costs for clinics and hospitals.EN

Researchers using shorter-wavelength ultraviolet lasers achieved dramatically more accurate chemical composition readings in industrial coating analysis. The advance matters: manufacturers relying on precise material characterization for semiconductors, coatings, and alloys can now reduce costly measurement errors and accelerate product development cycles.EN

Researchers have formalized the discrete analogue of the Gaussian distribution, a function central to statistics, physics, and machine learning. The discovery unlocks new applications in differential privacy—critical for protecting user data in AI systems—and provides mathematical tools for modeling complex systems from quantum physics to network analysis.EN

Researchers have developed a mathematical framework to classify and predict when quantum systems fail in specific ways, solving a decades-old problem in physics. The discovery could accelerate development of more stable quantum computers and advanced materials by helping engineers design systems that harness these critical points rather than avoid them.EN

A new review of transfer line balancing research reveals the field has lacked a unified framework—until now. For automakers and aerospace manufacturers running high-volume production, this classification system could unlock efficiency gains by clarifying which optimization approaches work best for different factory configurations.EN

Engineers have developed a computational tool that can predict how nearby deep excavations will destabilize existing subway and utility tunnels—a critical safety concern in dense urban areas. The research offers a practical way for cities and developers to assess construction risks before work begins, potentially preventing costly damage and service disruptions.EN

Researchers have developed a way to grow dense titanium boride coatings—used to protect cutting tools—without heating substrates, reducing energy consumption during manufacturing. The technique uses ion bombardment instead of heat, offering manufacturers a path to lower production costs while meeting sustainability goals.EN

Researchers have developed a mathematical framework to predict when vehicles lose wireless connections during spectrum handoffs in next-generation 6G networks. The work addresses a critical reliability gap for autonomous vehicles and connected car services, offering telecom operators and automotive manufacturers a tool to minimize dropped connections in mixed licensed and unlicensed spectrum environments.EN

Researchers found that lightweight concrete made from recycled materials can insulate sound as effectively as traditional concrete while reducing environmental impact. The discovery matters for social housing developers seeking to cut carbon emissions without compromising building quality or resident comfort.EN

Researchers have demonstrated a hybrid optical-fiber-satellite link that enables high-speed data transmission from underwater sensors to surface networks—a capability that could unlock deep-sea resource monitoring and offshore infrastructure surveillance. By converting visible light signals to infrared at the water's surface, the system maintains data speeds of 8 gigabits per second across varying ocean conditions.EN

Researchers have cracked a long-standing computational bottleneck that has hindered AI systems and software verification tools. The breakthrough, implemented in a tool called POQER, uses adaptive approximation techniques to solve previously intractable mathematical problems, potentially accelerating formal verification for autonomous systems and financial software where safety guarantees are critical.EN

Scientists have engineered a new class of high-entropy alloys that combine multiple metals into stable, magnetic materials capable of withstanding extreme temperatures. The breakthrough uses a rapid processing technique that could accelerate production of stronger, lighter components for aerospace, automotive, and defense applications.EN

Researchers have developed a machine learning system that can accurately estimate battery lifespan and degradation despite missing or corrupted sensor data—a critical capability for electric vehicles and grid storage systems. The method maintains accuracy even when half the data is missing, potentially reducing costly premature battery replacements and improving the economics of EV manufacturing and energy infrastructure.EN

Researchers have identified why silver-based solar absorbers behave differently from copper versions—a finding that could unlock efficiency gains in next-generation photovoltaic technology. The discovery, using advanced spectroscopy, reveals how atomic arrangements affect material performance in cells already reaching 23% efficiency, pointing manufacturers toward design improvements.EN

Researchers have mapped how competing respiratory disease strains interact within populations, finding that quarantine effectiveness determines which strain survives—but imperfect isolation measures can keep both circulating indefinitely. The findings matter for public health agencies designing containment strategies and companies planning workforce policies during outbreaks.EN

Researchers have developed a faster way to identify and neutralize compromised sensors in automated control systems—a common vulnerability in industrial plants, power grids, and autonomous vehicles. The breakthrough reduces computational demands by 80-90%, making real-time defense feasible even as systems scale up with more sensors.EN

Researchers have developed a computational method that preserves the fundamental laws of physics during numerical simulations while automatically adjusting resolution where needed. The advance could accelerate engineering design cycles and reduce computational costs for industries relying on complex physics modeling, from aerospace to energy infrastructure.EN