"Seesaw magnetoelectricity makes low-energy switching easier without weakening the magnetic state, thereby keeping the readout signal strong," Zhang said. "It provides clear 'yes' or 'no' signals, not a 'maybe.'"

Computer simulations show the approach boosts the switching rate from under 100 flips per second to over 400,000, ultimately generating more than a million random bits every second. The output passed National Institute of Standards and Technology benchmark suites — the standard yardstick for randomness in this area.

Who built the blueprint?

Zhang and Boris Yakobson, the Karl F. Hasselmann Professor in Engineering at Rice, co-led the study. The team collaborated with Shuai Dong, chair of the School of Physics at Southeast University in China. Yakobson said the project grew out of broader work on how physical behavior produces entropy — a measure of disorder that, here, translates directly into random bits.

"Our broader interest in TRNG, or how to extract entropy from physical behavior and convert it into random bits, focused mostly on charge-fluctuating entities in field-effect transistors," Yakobson said. "But when exploring autferroics' energy landscape, especially the lower barrier separating opposite polarizations, Jun-Jie proposed this might lead to faster TRNG. It turned into a very fruitful collaboration with our recent report dovetailing with our previous one."

Why four states, not two?

Standard digital bits hold a 0 or a 1. Autferroics can hold four distinct, equally stable combinations of electric and magnetic states. One tiny component could do the work of several conventional transistors.

The team showed that wiring two such units to simple circuits lets ordinary microchips process multi-state information and mimic quantum-style computation by testing several answers at once. Beyond encryption, this offers a secondary path toward future microchip architecture.

"Four-state autferroic devices can represent more information in a single device," Zhang said. "For example, these multistates allow complex numbers, a combination of a real and an imaginary number, to be encoded directly in hardware."

What's still missing?

The findings rest on computer models of a 2D nanomaterial called titanium germanium selenide, not on a physical prototype. Structural imperfections in real devices may slow switching speeds somewhat. The team describes the work as a theoretical blueprint rather than a finished chip.

"This research could be useful for computing, data encryption and processing and other information technologies," Zhang said. "It aims to make true random number generators faster and more reliable."

The researchers also found that applying a constant electric field accelerates switching further. Because the field pushes both directions equally, the device stays at a perfect 50/50 balance — no bias toward 0 or 1, hence the "seesaw" name.

Details appear in: Jun-Jie Zhang et al, "Autferroics-Based True Random Number Generators with Enhanced Performance," Physical Review Letters (2026). DOI: 10.1103/8vd7-ffng. A preprint is available on arXiv at arxiv.org/abs/2608.13838.