Scientists just built an AI where the 'neurons' are made of a single electron or a single particle of light.
April 17, 2026
Original Paper
Training single-electron and single-photon stochastic physical neural networks
arXiv · 2604.10861
The Takeaway
Modern AI runs on massive chips that simulate how a brain works. This team did the opposite: they used the actual, physical behavior of single particles to do the thinking. They trained neural networks where a single electron jumping through a gap acts as a logic gate. It is the ultimate hardware AI, where the laws of physics are the software. This could lead to computers that use millions of times less power than today's GPUs while being inherently faster.
From the abstract
The computational demands of deep learning motivate the investigation of alternative approaches to computation. One alternative is physical neural networks~(PNNs), in which learning and inference are performed directly via physical processes. Stochastic PNNs arise when the underlying neurons are realized by the dynamics of a stochastic activation switch. Here we propose novel electronic and photonic stochastic neurons. The electronic realization is implemented by single-electron tunneling throug