Tech & AI

Researchers have created the first detailed measurements of blast waves from underwater electrical explosions, capturing pressure dynamics invisible to conventional tools. The work bridges experiment and simulation to reveal energy transfer mechanisms—findings with potential applications in naval engineering, offshore safety, and underwater demolition industries.EN

Researchers have designed a shared data platform for tracking and reusing precast concrete components across multiple countries. The system addresses a critical gap in the circular construction industry, where fragmented data storage has prevented companies from scaling concrete recycling operations and capturing cost savings.EN

Researchers have identified geometry modifications that prevent sudden pitch-up failure in delta-wing aircraft at extreme angles of attack—a critical safety hazard. The fix extends safe flying range without sacrificing speed, potentially improving performance and safety for military and aerospace applications.EN

Researchers have solved a decades-old problem in optimization theory: when algorithms designed to find solutions encounter impossible constraints, they can still extract valuable information about what's actually achievable. The discovery has implications for engineering design, resource allocation, and any field that needs to work around conflicting real-world demands.EN

Scientists have engineered organic solar panels that remain stable at temperatures above 200°C—a breakthrough that could extend device lifespan and reduce manufacturing costs. The advance uses a fullerene alloy blend that maintains performance during high-heat processing, addressing a major obstacle preventing organic photovoltaics from competing with silicon-based solar technology.EN

Researchers have solved a longstanding technical problem in wireless communication by developing formulas that accurately predict how networks perform under realistic signal conditions. The breakthrough could help telecom companies and device makers design faster, more reliable networks for latency-sensitive applications like autonomous vehicles and remote surgery.EN

Researchers have developed a complete mathematical framework for categorizing higher-weight Hankel forms—abstract operators used in signal processing and quantum mechanics. The breakthrough could simplify how engineers and physicists classify and work with complex systems that process multidimensional data.EN

Scientists have detailed how benzene rings function as building blocks in conjugated polymers, materials that behave like semiconductors and can be engineered into conductors. The findings could accelerate development of cheaper, more efficient organic electronics for displays, transistors, and next-generation devices.EN

New modeling shows that renewable energy policies and market changes across Europe—not just in the Nordic region—will significantly alter Nordic electricity prices, trade flows, and grid stability. The finding underscores why energy companies and policymakers must coordinate cross-border strategies to manage costs and market risks in an increasingly integrated European grid.EN

A new study maps how digital creators in Armenia, Georgia, Moldova, and Ukraine are abandoning traditional advertising revenue to use crowdfunding and subscription tools like Patreon and GoFundMe. As Facebook and Google capture most ad dollars, this shift signals how creators worldwide are rebuilding sustainable income independent of platform gatekeepers.EN

Swedish researchers used the country's centuries-old population registry to verify 28 people who lived to 110 or beyond—a rare feat of demographic documentation. The finding underscores the value of robust government data systems for health research, insurance modeling, and understanding human longevity limits.EN

Researchers have developed a smarter way to monitor multiple related data streams over shared wireless networks — cutting unnecessary transmissions while keeping tracking errors minimal. The technique could improve real-time monitoring systems in industrial IoT, autonomous vehicles, and remote sensing, where bandwidth and latency are costly constraints.EN

Scientists have cracked a decades-old problem in computational modeling: keeping complex simulations stable when data flows in from unpredictable external sources. The breakthrough uses a new mathematical framework that automatically prevents solutions from spiraling out of control, which matters for industries relying on climate forecasts, fluid dynamics simulations, and any real-time model processing live sensor feeds.EN

Linköping University published a collection of seven applied mechanics projects completed by students working directly with industry partners in 2025. The coursework addressed real-world challenges in structural design, fluid dynamics, and optimization—areas critical to manufacturing, aerospace, and energy sectors seeking practical solutions to complex engineering problems.EN

Researchers have created TinySEED, a compact artificial intelligence system that detects sophisticated DDoS attacks faster and more accurately than existing methods. The breakthrough matters because it allows businesses to protect network infrastructure at the edge—where most attacks occur—without the expensive computing power that traditional defenses demand.EN

Researchers have discovered a novel geometric framework that simplifies how scientists understand complex algebraic structures called semifields—with potential applications in cryptography and quantum computing. The breakthrough emerged from studying how tensors behave over finite fields, revealing unexpected connections to quantum-like geometric surfaces that could reshape how engineers design secure systems.EN

Researchers have developed a technique that extracts useful information from signals that sensors fail to detect, dramatically improving location-finding accuracy in weak signal environments. The breakthrough, tested on Bluetooth devices, could boost performance for wireless networks, autonomous systems, and indoor positioning—industries increasingly dependent on reliable signal direction sensing.EN

Researchers have solved longstanding mathematical problems that underpin how engineers filter and process signals in telecommunications, audio processing, and radar. The breakthrough makes it faster and more reliable to design systems that remove noise from data — a critical step in everything from 5G networks to medical imaging.EN

A new paper argues that AI agents powered by large language models are fundamentally unreliable without mathematical verification systems. Since these models hallucinate and give inconsistent answers, companies deploying them in high-stakes applications face serious liability and safety risks unless they adopt formal verification methods.EN

Researchers have released an updated web-based software called ViNCent 2.0 that identifies the most influential nodes in large, fast-changing networks — a capability companies need to optimize supply chains, detect fraud, and understand social influence. The tool handles multiple importance metrics simultaneously, making it practical for real-world data that includes time changes and multiple data types.EN

A major analysis across three countries reveals a persistent gender gap in mental rotation—a spatial ability critical for STEM success. The findings could reshape how schools identify and train students for science and tech careers, where spatial reasoning directly predicts academic performance and career choice.EN

A new study finds that logistics companies replacing diesel trucks with electric ones without rethinking operations are leaving money on the table. By optimizing routes, fleet composition, and charging schedules together, operators can cut costs twice as much and electrify 85% of deliveries instead of 48%—showing that the business case for e-trucks hinges on operational redesign, not just hardware swaps.EN

Researchers have published a correction to a 2022 computational method that makes fluid dynamics simulations more accurate and stable. The fix matters for companies and agencies relying on weather prediction, climate modeling, and engineering simulations—anywhere computational models of flowing gases and liquids drive critical decisions.EN

Researchers have trained neural networks to simulate expensive 32-microphone audio arrays using just 4 microphones, slashing hardware costs for acoustic monitoring systems. The breakthrough could accelerate adoption of sound-based quality control, surveillance, and environmental monitoring across manufacturing, smart buildings, and industrial facilities.EN

A critical analysis of the DE-BIAS project shows that while automated systems can flag harmful language in digitized cultural collections, they struggle with context and interpretation—raising questions about who decides what's biased. The finding matters as institutions worldwide rush to deploy AI for heritage preservation without fully understanding the technology's blind spots.EN