Electronic skin (e-skin), a stretchable, conformable, and multimodal sensing platform, is rapidly advancing in healthcare, robotics, and human–machine interaction (HMI). However, prolonged wear and sweat accumulation at the e-skin-skin interface can introduce signal artifacts, cause discomfort, and eventually lead to interfacial failure. Inspired by the unidirectional liquid transport of the Nepenthes peristome, we constructed an intelligent breathing e-skin (SPTL) featuring a Janus bilayer structure that creates a “liquid diode” effect for active sweat transport, achieving a remarkable cumulative one-way transport index of 956.36 while maintaining a dry interface even under profuse perspiration. This device exhibits a high tensile strain of 627%, a breathability of 20.02 mm s⁻¹, and a pressure sensitivity of 7.39 kPa⁻¹. Furthermore, the SPTL demonstrates versatile multimodal sensing capabilities, including real-time Morse code and non-contact capacitive sensing, while sustaining stable performance over 10,000 pressing cycles (capacitance decay < 3%). As a bio-integrated electrode, SPTL enables high-fidelity acquisition of electroencephalogram (EEG), electromyogram (EMG) and electrocardiogram (ECG) signals, significantly outperforming commercial Ag/AgCl electrodes. Integrated with machine learning algorithms and EMG signals captured by the SPTL, it facilitates precise teleoperation of a quadruped robot and handwritten letter recognition with over 95% accuracy. This bionic strategy provides a versatile solution for intelligent, breathable, and multimodal bio-integrated interfaces.

The human brain performs perception, memory, learning and decision-making with only about 20 watts of power. A new study in National Science Review proposes a physics-based framework to explain this extraordinary efficiency. By modeling cooperative neuronal activity, the researchers suggest that brain information may be stored and processed through dynamic electrical patterns, offering new clues for brain science and energy-efficient artificial intelligence.

Seoul National University College of Engineering announced that a research team led by Professor Kyunghan Lee of the Department of Electrical and Computer Engineering has developed an artificial intelligence system, “Ouroboros,” that performs Vision Transformer-based video analysis 2.61 times faster with 13.0% computation than conventional methods on edge devices.

Seoul National University College of Engineering announced that a research team led by Prof. Hyun Oh Song of the Department of Computer Science and Engineering, in collaboration with Samsung AI Center, has developed “Rule2DRC,” an AI agent technology that automatically generates inspection codes required for semiconductor design verification.

A stretchable sensor made from self-healing materials and liquid metal detects touch, senses damage and repairs itself without external power, potentially extending the working life of underwater machines used in diving and marine robotics.