An AI chip can be completely rewired using nothing but a radio broadcast signal.
April 29, 2026
Original Paper
Remotely programming the weights of a spintronic neural network by a radiofrequency broadcast signal
arXiv · 2604.24561
The Takeaway
Spintronic neural networks can now have their synaptic weights programmed remotely using radiofrequency signals. This breakthrough eliminates the need for complex, individual wiring to every single connection on a physical AI chip. By broadcasting a specific signal, researchers can tune the hardware synapses to learn new tasks or optimize their performance. This makes the hardware much easier to scale up to millions or billions of connections, similar to the human brain. It opens the door to brain-like computers that can be updated or reprogrammed wirelessly while they are embedded inside robots or other machines.
From the abstract
Selectively programming large number of non-volatile synaptic weights without compromising scalability is a key challenge for in-memory computing. Here, we demonstrate remote programming of synaptic weights in series-connected chains of 11 vortex-based magnetic tunnel junctions using broadcast radiofrequency signals applied through a shared strip line. The programming relies on frequency-selective reversal of the vortex-core polarity and therefore does not require individual access lines or sele