China starts mass production of world’s first non-binary AI chip in 2025
China starts mass production of world’s first non-binary AI chip overcoming traditional computing barriers and will be used in touch displays, flight systems and aircraft navigation
As per the report by the South Morning China Post, China has initiated the world’s first large-scale application of non-binary AI chips, integrating its proprietary hybrid computing technology into critical sectors including aviation and industrial systems.
Spearheaded by Professor Li Hongge’s team at Beihang University in Beijing, this breakthrough overcomes fundamental barriers in traditional computing by merging binary and stochastic logic, enabling unprecedented fault tolerance and power efficiency in intelligent control applications like touch displays and flight systems while sidestepping US chip restrictions.
Today’s chip technologies face two big challenges: the power wall and the architecture wall, Li told the Beijing-based official newspaper Guangming Daily last month.
The power wall stems from a fundamental contradiction – while binary systems are efficient at carrying information, they consume a large amount of power. The architecture wall is caused by the fact that new non-silicon chips cannot easily communicate with traditional systems based on CMOS or complementary metal-oxide-semiconductors.
Li’s team had been exploring alternatives since 2022. Their breakthrough came with the proposal of a new numerical system – Hybrid Stochastic Number (HSN) – which combines traditional binary numbers with stochastic or probability-based numbers.
Binary logic, the foundation of today’s computing, represents variables using 0s and 1s and relies on precise arithmetic operations. However, large-scale binary computations require extensive hardware resources.
The HSN system aims to reduce the hardware load and improve performance by introducing a form of probabilistic computation that is more flexible and tolerant to minor errors. This is especially useful in scenarios where absolute precision is not required for every operation, such as image processing, environmental sensing, and embedded control systems.
By designing the non-binary AI chip architecture around the HSN model, Li’s team was able to minimize the transistor count needed for computations, thereby reducing power consumption. This makes the chips not only more efficient but also more scalable in deployment, particularly in sectors that demand high energy efficiency and real-time processing.
The chip’s compatibility with existing CMOS-based systems also marks a major technical achievement. Rather than requiring an entirely new infrastructure, the non-binary chip can operate within legacy environments, easing its integration into aircraft systems, industrial robots, and advanced touch interfaces.
This compatibility is especially important in China’s current semiconductor landscape, which has been shaped by ongoing tensions with the United States and restrictions on advanced chip imports. Developing domestically viable and export-restricted-free chip solutions like this allows China to reduce its reliance on foreign technology and assert greater control over its tech ecosystem.
According to the report, the non-binary chips have entered mass production and are being actively deployed across use cases where resilience and adaptability are critical, such as autonomous vehicle controls, aerospace navigation, and rugged industrial environments.
The team’s work has also garnered interest in academic and commercial research circles, with potential collaborations forming to adapt the HSN framework to other computing models, including neuromorphic and edge AI platforms.
While traditional chips continue to serve the core infrastructure of data centers and consumer electronics, the rise of non-binary alternatives could signal a shift toward more application-specific architectures—especially in fields that benefit from probabilistic and adaptive logic systems.
If successful at scale, China’s early leadership in this non-binary AI chip domain could give it a technological edge not only in emerging industries but also in academic computing research and national infrastructure.
About Beihang University’s Initiative
This initiative is led by Beihang University’s School of Computer Science and Engineering in partnership with domestic chip manufacturers. It is supported by China’s broader push for semiconductor independence and is aligned with national strategies to reduce reliance on imported technologies in critical sectors such as aerospace, industrial automation, and AI infrastructure.
More info here – Have a Story? Address it to the Editor and submit it here
Disclaimer
The information provided in this article is for general informational purposes only and from publicly available sources. While we strive for accuracy, we do not make any representations or warranties, express or implied, regarding the completeness, reliability, or validity of the content. This article does not make any direct claims about specific companies, individuals, or organizations. Any references to reports or external sources are for context and do not imply endorsement or verification of any specific allegations. Readers are encouraged to conduct their own research and seek professional advice before making business decisions. We disclaim any liability for any losses or damages incurred as a result of reliance on the information provided.