“Thermodynamic computers” can mimic AI neural networks – and use orders of magnitude less energy to produce images

Scientists have built a “thermodynamic computer” that can generate images from random data disturbances, i.e. noise. In doing so, they imitated the generative artificial intelligence Neural Network (AI) Capabilities – Collections of machine learning algorithms modeled on the brain.

Above absolute zero temperatures, the world hums with energy fluctuations called thermal noise, which manifest themselves in wiggling atoms and molecules, changes in direction on the atomic scale for the quantum property that gives magnetism, and so on.