Solid‑state hydrogen storage offers high density and safety, but material development is slow. A comprehensive review highlights how machine learning (ML) and big data are accelerating discovery. The Digital Hydrogen‑S platform integrates over 3000 materials and 254,000 structured records, including multimodal data (PCT curves, kinetic data, synthesis metadata). Analysis shows that nearly no alloy meets US DOE targets for capacity, temperature, and pressure. ML models now predict storage capacity, formation enthalpy, and pressure‑composition‑temperature curves with high accuracy. Neural network potentials (NNPs) provide near‑first‑principles insight into hydrogen adsorption, dissociation, and diffusion. Future priorities include open‑access multimodal databases, inverse design, and generalized NNPs for complete absorption‑desorption cycles.

EPFL researchers have used a neural network to create all-atom models of proteins, as well as the dynamic movements that govern their function. Their comprehensive yet simplified approach resolves a major bottleneck in biology.

Researchers at the University of Pennsylvania have developed ApexGO, an AI-powered method for improving promising but imperfect antibiotic candidates. Unlike many AI approaches that screen databases for molecules that might work, ApexGO starts with a small number of starter molecules and makes step-by-step edits, using a predictive algorithm to guide the search toward stronger versions. In tests against disease-causing bacteria, 85% of the AI-generated molecules halted bacterial growth, while 72% outperformed the peptides from which they were derived. The work points toward a more systematic approach to antibiotic discovery, using AI to guide researchers toward molecules most worth testing.

A technology that surpasses the limitations of existing sensors, which failed to distinguish between water and asphalt on dark roads, has emerged to enhance the accuracy of autonomous driving and medical diagnostics. Our university's research team has developed a next-generation polarization sensor that can read the "direction" of light and change its own response. KAIST announced on May 12th that a research team led by Professor Joonki Suh from the Department of Chemical and Biomolecular Engineering has developed a "self-reconfigurable" polarization sensor array technology that regulates its operation by finding the optimal state using "polarization" information—the property of light vibrating in a specific direction. With the recent explosive increase in data and the rapid development of artificial intelligence technology, the need for next-generation vision systems that can efficiently process vast amounts of information with low energy is growing. However, existing image sensors only detect the intensity (brightness) of light, limiting their ability to precisely grasp the orientation or surface structure of objects. To overcome these limitations, the research team developed a polarization-based sensor technology capable of recognizing the vibration direction of light. In particular, by utilizing a "heterostructure" that combines two different materials—tellurium (Te) and rhenium disulfide (ReS₂)—they effectively implemented characteristics where the response to light varies depending on the crystal orientation.

The Okinawa Institute of Science and Technology (OIST) has opened applications for the 2027–2028 cohort of its flagship Innovation Accelerator. The 10-month program supports entrepreneurial researchers developing deep technologies with strong potential for commercialization in the Japanese market.