Cisco unveils integrated AI foundation with security and performance focus amid enterprise shift

Cisco launches its Secure AI Factory with NVIDIA, offering a modular, security-rich architecture designed to streamline large-scale AI deployment and support business resilience in Asia-Pacific and Japan.

Cisco is pitching a more integrated answer to a problem that is becoming familiar in boardrooms across Asia-Pacific and Japan: how to move from AI experimentation to dependable, large-scale deployment without exposing the business to security and performance failures. In a post published on 27 April, Cisco argued that many organisations have already moved beyond the question of whether to use AI and are now focused on whether their infrastructure can support it securely and consistently.

The company says the answer lies in a purpose-built foundation rather than a patchwork of tools. Its Secure AI Factory with NVIDIA is presented as a modular reference architecture that brings together compute, networking, storage, software and security in one design. Cisco says the aim is to reduce the integration problems that often accompany enterprise AI programmes and to give customers a platform that can be deployed with less friction across the core and the edge.

Cisco has also been stressing the security side of that proposition. According to the company’s product material, the platform embeds protections across the stack through tools including Cisco AI Defense, Hybrid Mesh Firewall, Hypershield and Splunk Enterprise Security. Cisco says that approach is intended to make security continuous rather than an afterthought, especially as businesses move AI workloads into production environments where data governance and operational resilience matter as much as raw model performance.

Performance remains the other half of the pitch. Cisco says the architecture uses Ethernet networking powered by Cisco Silicon One and validated with NVIDIA Spectrum-X, while its AI PODs combine Cisco UCS servers with storage from partners including Pure Storage, NetApp, Hitachi Vantara and VAST Data. In Cisco’s telling, the point is to give enterprises enough bandwidth and scale for demanding AI workloads without forcing them to abandon familiar operational models.

The company is also tying the technology offer to a broader partner strategy. In January, Cisco launched its 360 Partner Program, which replaces older tier labels with designations based on validated capability, including specialisations aimed at AI-ready infrastructure. Cisco says customers should look for partners with the right technical credentials, relevant industry knowledge and the ability to support deployments well beyond initial go-live, as model drift and changing workloads can make AI programmes a long-term commitment rather than a one-off project.

Source Reference Map

Inspired by headline at: ^[1]

Sources by paragraph:

Source: Fuse Wire Services