Topics

By using plain language to describe tasks, the system automatically selects models and hardware to save energy and money.

Scientists at Oregon State University have built a single device that detects light, stores information, and adjusts memory strength electronically, offering a path to faster and lower-power artificial intelligence systems.

A new hair-thin chip uses light to transfer massive data quickly, saving energy for artificial intelligence systems.

As a new UN report declares a global water bankruptcy crisis, the AI industry's massive resource demands collide with Earth's limits, forcing a pivotal turn toward space-based infrastructure.

Solar-powered satellite networks promise unlimited energy for artificial intelligence, enabling massive computing scale without earth's resource demands.

Researchers develop a new type of neural network that learns continually with much lower power consumption than today's large AI models.

UC San Diego researchers create a computing device that mimics brain networks to combine memory and processing for faster and more energy-efficient performance.

Engineers at UC San Diego have created a better way to teach AI systems to handle tough problems mixing words and pictures, like math questions with charts.

Researchers introduce recursive language models that help AI handle very long texts without losing accuracy, using self-calls and code environments to break down and analyze information efficiently.

This innovation separates core logic from search methods in AI systems, making it easier to fix mistakes and improve performance without major code changes.