Currently, mainstream dynamic random-access memory (DRAM) consumes approximately 200,000 electrons to store 1 bit of information. The latest research from Fudan University's Zhou Peng and Liu Chunsen, published in Science magazine on July 17, offers an alternative path: a single electron can accomplish the same storage task. Dubbed 'quantum flash memory,' this technology has, for the first time, observed stable single-electron storage behavior at room temperature (27°C), with a 0.5-volt storage window meeting commercialization standards.

This achievement approaches a fundamental physical limit. Theoretically, 'one electron, one bit' represents the ceiling for storage density. However, due to the minuscule charge of a single electron, observing its storage state is akin to perceiving the ripple of a single droplet in a vast reservoir. In previous experiments, single-electron states struggled to remain stable at room temperature, quantum behavior remained unclear, and theoretical support for engineering non-volatile quantum storage was lacking.

Building on the fundamental principles of quantum mechanics and leveraging the electron confinement advantages of two-dimensional semiconductors at the atomic level, Zhou Peng and Liu Chunsen's team designed a quantum flash memory device named 'Guiyi.' This device achieves a 0.5-volt storage window by injecting just a single electron, reaching the theoretical peak of 'one electron, one bit' charge storage. Notably, the team also introduced the innovative 'State Density Scissors' theory, revealing for the first time in energy space an anomalous quantum storage behavior: specific quantum states are precisely 'cut' and vanish into thin air.

The research team described this mechanism as using an invisible 'quantum scissors' in energy space to precisely trim specific quantum states, causing them to disappear. This is the first anomalous quantum storage behavior ever observed worldwide.

A more critical implication is that data interaction latency and power consumption issues caused by memory chips represent the fundamental bottleneck hindering computing power advancement. As the foundational cornerstone of computing power, the energy efficiency of memory devices directly determines the overall performance ceiling of artificial intelligence systems. Quantum flash memory's ability to reduce computing power consumption from the ground up makes it a core candidate solution for next-generation memory aligned with the development of artificial general intelligence.

This achievement fills a critical theoretical gap in the engineering application of quantum storage. The establishment of a single-electron quantum storage theoretical framework signifies that quantum storage is transitioning from the laboratory to large-scale application, completing a vital piece of the puzzle. For the rapidly scaling artificial intelligence industry, memory architecture innovation may prove more decisive than mere computing power accumulation. The leap from 200,000 electrons to a single electron represents not just an increase in storage density but also a redefinition of physics' boundaries at the fundamental level.

As quantum flash memory moves from theory to engineering, the foundational cornerstone of computing power will be relaid. The trigger for the next leap in artificial intelligence computing power may lie not in the computing chips themselves but in whether memory devices can break through this physical threshold.