Cisco Systems has unveiled its Silicon One G300 switch chip and a new high-performance router, marking a significant step in accelerating data movement across large-scale AI data centres. The launch positions Cisco to compete directly with established players in the rapidly expanding $600 billion global AI infrastructure investment cycle. 

Designed to support the growing demands of AI training and inference workloads, the Silicon One G300 enables AI systems to communicate efficiently across hundreds of thousands of interconnected links. The chip is expected to enter the market in the second half of the year, aligning with rising enterprise and hyperscaler investments in AI-driven computing environments. 

Manufactured using TSMC’s advanced 3-nanometer process technology, the G300 introduces multiple network-resilience features, including built-in “shock absorber” mechanisms. These capabilities help prevent network slowdowns during sudden surges in data traffic a common challenge in AI-heavy workloads. The chip can automatically reroute data around congestion or faults within microseconds, maintaining performance consistency at scale. 

According to Cisco, deployments powered by Silicon One G300 can deliver up to 28% faster completion of certain AI computing tasks, improving overall system efficiency and utilization. This performance gain is critical for organizations running high-density AI clusters where latency, reliability, and throughput directly impact operational costs. 

With this launch, Cisco strengthens its role in next-generation data centre networking, offering enterprises and cloud providers a pathway to build faster, more adaptive, and AI-optimized infrastructure capable of supporting future workloads. 

To find out more, feel free to browse our latest updated relevant report:

https://semiconductorinsight.com/report/artificial-intelligence-ai-chips-market/