Comment

Somatosensory feedback is an essential feature of neural prostheses that aim to restore natural hand dexterity after neurological injuries. This Comment discusses the potential and challenges of electrically stimulating sensory neurons, highlighting the need for highly selective neurotechnology and biomimetic-encoding algorithms for effective brain–machine interfacing.

The endurance, retention and system-level performance of memristors for memory and computation has been often misrepresented in articles that lack statistics and use non-standardized characterization and simulation protocols. Here we discuss the origin of these issues, their negative effect in the nascent memristor industry, and potential ways to mitigate them.

Learning and embodiment are intertwined, resulting in a mutually reinforcing effect. Research should aim not only for learning to enhance embodiment, but also, more importantly, for embodiment to facilitate learning. Achieving synergy between these two aspects remains an ongoing challenge.

Designing marketable 6G measurement tools presents technical and logistical challenges, including generating and capturing signals in new spectrum allocations with high speed and low latency, managing wide bandwidths, and ensuring system-level synchronization. Overcoming these challenges requires advances in hardware design and implementation of instrumentation tools tailored for 6G test environments.

Wearable bioelectronics are reshaping patient care through seamless, personalized monitoring. As technology is gradually becoming smoothly integrated into our daily life, its success will depend not only on innovation, but also on our ability to ensure responsible integration — balancing performance, usability and equitable access.

As digitalization permeates societal interactions to connect people over physical distances, mediated social touch (haptics in human-to-human interaction) remains overlooked despite its high potential in times of social divides and mobility. Disciplinary silos must be broken, and an intercultural approach is needed to integrate touch into digital global societies.

Autonomous vehicles rely on both LiDAR and cameras for perception, with each technology offering unique advantages — cameras provide rich contextual information, whereas LiDAR delivers precise depth data. Understanding their trade-offs is crucial for creating reliable and efficient autonomous vehicles.

Optical wavefront shaping compensates distortions caused by scattering, aberrations or inhomogeneities in optical medium, enabling precise phase control to enhance light penetration in turbid environments. The integrated phase measurement sensor combines light sensing and modulation at pixel level within a single device, thereby reducing alignment constraints and bandwidth limitations.

Ungendering engineering is necessary to break the tight connection between masculinity and engineering and make it a more inclusive profession. To attract more women to engineering, the gendered stereotypes attached to engineering must change.

Agentic artificial intelligence (AI) workflows, integrated with 6G-enabled edge computing, present a transformative approach to reduce hallucinations in generative AI. Combining real-time tool integration, self-reflection and collaborative multi-agent systems, this framework offers scalable, reliable solutions for diverse applications, advancing the accuracy and adaptability of generative AI in dynamic and data-intensive environments.

Amid geopolitical crises and economic challenges, it is crucial to reassess the future role of fossil fuel resources, particularly oil and gas fields. A promising opportunity now exists to repurpose these fields for geothermal energy, helping to accelerate the energy transition while mitigating the risk of resource stranding.

Female researchers in Japan face unique challenges caused by cultural and institutional norms, making their career progression more difficult than in other developed countries. Overcoming these barriers and fostering an inclusive academic environment require a systematic approach that includes targeted interventions and clear, enforceable measures to ensure compliance and accountability.

Traditional DRAM technology, with memory bit cells consisting of one silicon transistor and one capacitor, faces major scaling challenges. A new DRAM bit cell without a capacitor and with two thin-film transistors — each with an oxide semiconductor channel such as indium-gallium-zinc-oxide — shows promises for continuing the DRAM technology roadmap, clearing the way for high-density 3D DRAM.

Bridging the divide between mathematical innovations and real-world applications is essential for addressing global challenges. By fostering interdisciplinary collaboration, diversity and inclusion, we can unlock the full potential of mathematical findings, driving innovation across applied disciplines and delivering mathematical solutions to society’s pressing problems.

Aqueous sulfur-based redox flow batteries (SRFBs) are promising candidates for large-scale energy storage, yet the gap between the required and currently achievable performance has plagued their practical applications. Here, we propose several engineering strategies towards SRFB commercialization.

Autonomous electric vehicles (AEVs) will soon share roads with traditional traffic participants — human-driven vehicles, pedestrians and cyclists — which will require careful planning for safe interactions. Here we advocate for the development of human-like driving technologies for AEVs through human-inspired approaches and the Turing test.

Soft robots offer an ideal embodiment for artificial intelligence integration owing to their inherent safe interaction with humans. However, their growth is impeded by low standardization. Origami-based standardized soft actuators and a LEGO-like assembly strategy enable the creation of complex, functional, soft robotic systems.

High-frequency energy conversion is essential in modern systems, with most relying on solid-state conductors. However, traditional materials such as copper and aluminium face challenges such as skin effects, proximity effects and inefficiency; exploring liquid metal composites could enable liquid–solid hybrid energy conversion, advancing applications such as power converters, motors, magnetic fields and rapid recycling.

The FAMES pilot line on fully depleted silicon-on-insulator technology provides a complete toolbox of technologies for the development of innovative chip architectures with increased performance and lower power consumption of mixed-signal circuits.

Grid-scale, long-duration energy storage has been widely recognized as an important means to address the intermittency of wind and solar power. This Comment explores the potential of using existing large-scale hydropower systems for long-duration and seasonal energy storage, highlighting technological challenges and future research directions for pumped-storage renovation.