Brain-Inspired Chip Could Slash AI Energy Use by 70%
Scientists have engineered a new nanoelectronic device using hafnium oxide that mimics how neurons work, operating with ultra-low power and potentially cutting AI energy consumption by up to 70%.
Breakthrough in Brain-Inspired Computing
Researchers have engineered a new nanoelectronic device using a modified form of hafnium oxide that mimics how neurons process and store information at the same time, operating with ultra-low power—potentially slashing energy use by up to 70%.
How It Works
The Cambridge research team developed a modified version of hafnium oxide that functions as a highly stable, low-energy 'memristor' by adding strontium and titanium and using a two-step growth process to create small electronic gates, known as 'p-n junctions', at the interfaces between layers, allowing the device to change its resistance by adjusting the energy barrier at these interfaces.
The Innovation
Scientists have created a new type of nanoelectronic device that could significantly reduce how much energy artificial intelligence systems consume, offering a more efficient alternative to today's power-hungry AI hardware.