Researchers at Seoul National University have developed a breakthrough artificial skin capable of simultaneously detecting temperature and pressure, mirroring human tactile perception. Led by Professor Seung Hwan Ko in the Department of Mechanical Engineering, the team published their findings in Nature Materials in 2025, introducing a single ultrathin multimodal sensor designed to advance Physical AI systems. Traditional multimodal tactile devices typically rely on stacking multiple functional layers, resulting in bulky structures, slower response times, and complex signal processing. To overcome these limitations, the SNU team engineered a core-shell nanowire network composed of a silver core and a copper oxide shell. This single-layer architecture dynamically switches between thermal and mechanical sensing modes at a rate of sixteen times per second. The design achieves sub-microsecond response times for mechanical stimuli and millisecond-level detection for thermal changes, all within a unified platform. When integrated with a wireless switching board and trained AI models, the sensor demonstrated significant performance gains. By processing interleaved thermal and mechanical signals, the AI classification accuracy for identifying objects rose from approximately sixty-five percent using single-mode data to ninety-five percent. In practical trials involving a fingertip-mounted sensor, the system achieved an eighty-three percent accuracy rate across twenty everyday objects. Furthermore, the team validated a multi-array configuration capable of mapping temperature and pressure distributions with spatial resolution comparable to human skin, confirming the technology’s scalability from single-point detection to full artificial skin systems. The innovation addresses a critical requirement for Physical AI, a rapidly emerging field focused on enabling machines to interact with and make decisions based on real-world physical environments. High-fidelity tactile perception is essential for robots to safely manipulate objects, navigate complex spaces, and replicate human-like dexterity. By consolidating dual sensory inputs into one flexible, conformally attachable layer, the new platform significantly simplifies system architecture while enhancing sensing precision. Development was spearheaded by co-first authors Kyun Kyu Kim, now a researcher at Apple, and Junhyuk Bang, a postdoctoral fellow at the California Institute of Technology. Professor Ko emphasized that this represents the first demonstration of unified thermal and mechanical stimulus processing without sensor stacking. The research team anticipates the technology will serve as a foundational component for next-generation robotics, prosthetics, wearable electronic skin, soft robotic actuators, and advanced human-machine interfaces. The streamlined design and human-comparable resolution position the artificial skin as a pivotal enabler for the next wave of tactile-enabled intelligent systems.