Category Technology/Electronics

All-optical chip achieves 100-fold speed boost over top-tier NVIDIA chips

A new optical chip for AI that is 100 times faster than traditional computer chips
The architecture of Light Generative chip (LightGen). Credit: Science (2025). DOI: 10.1126/science.adv7434

Scientists in China have unveiled a new AIchip called LightGen that is 100 times faster and 100 times more energy efficient than NVIDIA chips, the leading supplier of AI chips worldwide. Instead of using electricity to move information, this new optical chip relies on light to perform complex generative tasks.

Traditional general AI models, such as ChatGPT and Stable Diffusion, run on everyday silicon chips and require massive amounts of computing power and electricity, which can generate significant heat. For particularly complex tasks, these chips can struggle with the workload, resulting in slow processing times.

Mimicking the human brain
Photons of light are already kno...

Read More

Ultra-low power, fully biodegradable artificial synapse offers record-breaking memory

In Nature Communications, a research team affiliated with UNIST present a fully biodegradable, robust, and energy-efficient artificial synapse that holds great promise for sustainable neuromorphic technologies. Made entirely from eco-friendly materials sourced from nature—such as shells, beans, and plant fibers—this innovation could help address the growing problems of electronic waste and high energy use.

Traditional artificial synapses often struggle with high power consumption and limited lifespan. Led by Professor Hyunhyub Ko from the School of Energy and Chemical Engineering, the team aimed to address these issues by designing a device that mimics the brain’s synapses while being environmentally friendly.

How the biodegradable synapse works
The result is a layered struct...

Read More

New memristor-based converter boosts energy efficiency in AI hardware

Cross-institutional team develops new memristor-based converter
Challenges in CIM systems for neural network computation. Credit: Nature Communications (2025). DOI: 10.1038/s41467-025-65233-w

A cross-institutional team led by researchers from the Department of Electrical and Electronic Engineering (EEE), under the Faculty of Engineering at The University of Hong Kong (HKU), have achieved a major breakthrough in the field of artificial intelligence (AI) hardware by developing a new type of analog-to-digital converter (ADC) that uses innovative memristor technology. The work is published in Nature Communications.

Challenges with conventional AI hardware
Conventional AI accelerators face challenges because the essential components that convert analog signals into digital form are often bulky and power-consuming...

Read More

Room temperature electron behavior defies expectations, hinting at ultra-efficient electronics

Scientists have discovered a way to efficiently transfer electrical current through specific materials at room temperature, a finding that could revolutionize superconductivity and reshape energy preservation and generation.

The paper is published in the journal Physical Review Letters.

The much-sought-after breakthrough hinges on applying high pressure to certain materials, forcing their electrons closer together and unlocking extraordinary electronic behaviors.

“Our research explores how electrons interact inside solid materials—interactions that give rise to remarkable phenomena such as high-temperature superconductivity and charge-density waves (CDWs),” said Dr. Mahmoud Abdel-Hafiez, associate professor of physics at the University of Sharjah.

“These effects are impo...

Read More