Key Takeaways

  1. Cisco (NASDAQ: CSCO), SharonAI Holdings (NASDAQ: SHAZ), and NVIDIA launched Australia’s first Cisco Secure AI Factory on February 23, 2026. It is housed at NEXTDC’s S3 data centre in Sydney.
  2. The facility runs on 1,024 NVIDIA Blackwell Ultra GPUs, along with Cisco UCS servers, Nexus Hyperfabric networking, and VAST Data storage systems—and it keeps all processing on Australian soil.
  3. Sharon AI recently secured a US$500 million debt facility from USD.AI. In addition, it completed a US$100 million convertible note raise to fuel GPU infrastructure expansion across Asia-Pacific.​
  4. The initiative directly supports Australia’s National AI Plan 2025. This plan outlines nine priority actions including infrastructure expansion, investment attraction, and responsible AI governance.

Quick Recap

On February 23, 2026, Cisco and SharonAI Holdings — an Australian neocloud listed on NASDAQ under the ticker SHAZ — officially announced the launch of Australia’s first Cisco Secure AI Factory. NVIDIA collaborated on building the facility. NEXTDC hosts it at its S3 data centre in northern Sydney. The facility keeps all AI data processing and model training strictly within Australian borders. Cisco APAC announced it simultaneously through Cisco’s official newsroom, BusinessWire, and its social media channels.

Inside the Sovereign AI Stack: Architecture and Partnerships

The Cisco Secure AI Factory is not a single vendor product — it is a multi-partner, vertically integrated sovereign AI platform. The compute layer uses 1,024 NVIDIA Blackwell Ultra GPUs—NVIDIA’s latest-generation accelerators for large-scale AI training and inference—and pairs them with Cisco UCS AI servers. The Nexus One unified management plane manages the stack, and Nexus Hyperfabric delivers the high-bandwidth, low-latency networking that GPU clusters demand.

Storage is provided by VAST Data, whose flash-based systems are optimized for the high-throughput data ingestion that AI training pipelines require. The entire infrastructure sits inside NEXTDC’s $1 billion S3 facility in Artarmon, Sydney — a Tier IV certified data centre with 80 MW of capacity. Security is layered throughout the stack using Cisco Hypershield for AI workload protection and Cisco AI Defense for securing model development, deployment, and usage.

Sharon AI, the operating neocloud partner, will deliver tailored GPU-as-a-Service solutions on top of this infrastructure. Customers gain access to sandbox environments for proof-of-concept testing, with solutions targeting defence, healthcare, financial services, and government sectors. Sharon AI CEO James Manning confirmed plans for additional cluster deployments in 2026 and beyond. Also, there is an expansion path that could push installed capacity beyond 20,000 Blackwell units.

Australia’s Strategic Moment : Why Sovereign AI Matters Now?

This launch arrives at a pivotal inflection point. Australia’s federal government released its National AI Plan in December 2025, setting out a three-part strategy to build smart infrastructure, spread AI adoption benefits, and keep Australians safe through regulatory frameworks. The plan explicitly calls for domestic data sovereignty and on-shore AI processing. This is exactly what the Cisco Secure AI Factory delivers.

The geopolitical dimension is equally important. Enterprises and government agencies across Asia-Pacific increasingly face regulatory pressure to keep sensitive data within national borders. NVIDIA Country Manager Sudarsharn Ramachandran described sovereign AI infrastructure as “a critical resource” for the region. Meanwhile, hyperscalers like AWS, Microsoft Azure, and Google Cloud all operate Australian data centres but remain headquartered overseas. This raises data jurisdiction concerns for defence and critical infrastructure clients.

Australia’s broader infrastructure ecosystem is moving fast. NEXTDC signed a memorandum of understanding with OpenAI in December 2025 to develop a $7 billion hyperscale AI campus at its S7 site in Western Sydney. This is expected to deliver up to 550 MW of capacity. Dell Technologies is partnering with Monash University and Macquarie Data Centres on the MAVERIC sovereign supercomputer. The Cisco Secure AI Factory adds another layer to this emerging sovereign compute stack. Specifically, it targets enterprise and government customers who need production-grade AI infrastructure now, not in 2027.

Competitive Landscape and Comparison

The sovereign AI and neocloud infrastructure market in Australia is rapidly evolving. Sharon AI (via the Cisco Secure AI Factory) now competes directly with emerging local players like SCX.ai and the growing Macquarie Data Centres AI partnership with Dell Technologies.

Feature / MetricCisco Secure AI Factory (Sharon AI)SCX.ai Sovereign AI NodeMacquarie Data Centres + Dell AI Factory
Primary Compute1,024 NVIDIA Blackwell Ultra GPUs​SambaNova SN40L ASICs​Dell PowerEdge + NVIDIA GPUs (MAVERIC)​
Data Centre HostNEXTDC S3, Sydney (Tier IV, 80 MW)​Equinix SY5, Sydney​Macquarie IC3 Super West, Sydney​
Networking StackCisco Nexus Hyperfabric + Silicon One​Standard Equinix interconnect​Macquarie Telecom network backbone​
StorageVAST Data flash systems​Not disclosedDell PowerStore​
Security LayerCisco Hypershield + AI Defense​Sovereign data residency claims​IRAP, ISO 27001, DISP compliance​
Target WorkloadsAI training + inference (enterprise/govt)​AI inference only (10x efficiency claim)​AI training + research (university/govt)​
Sovereign ComplianceFull on-shore processing, SOCI-aligned​Full on-shore processing​Full on-shore, government-grade​
Financial Backing$500M USD.AI facility + $100M convertible note​UndisclosedDell + ASX-listed Macquarie Technology Group​
Expansion Plans20,000+ Blackwell GPUs planned​National node network in 2026​IC3 Super West + MAVERIC scaling​

The Cisco-Sharon AI partnership leads in raw GPU compute scale and enterprise-grade security integration. With 1,024 Blackwell Ultra GPUs and Cisco’s full networking and security stack, they provide a turnkey solution for large-scale AI training. SCX.ai differentiates on energy efficiency and inference-specific workloads, claiming 10x better energy performance through ASIC-based architecture. However, independent benchmarks remain scarce. Macquarie Data Centres is backed by its ASX-listed parent and a 24-year track record. This brings deep government compliance credentials and research partnerships that appeal to the public sector and academic institutions.

Sci-Tech Today’s Takeaway

I think this is a genuinely big deal — not just for Cisco’s enterprise ambitions, but for Australia’s entire AI trajectory. In my experience covering the APAC tech infrastructure space, sovereign AI has been more slogan than substance for the past two years. What makes the Cisco Secure AI Factory different is that it’s real and operational. It is also backed by a financial stack that signals serious scale: Sharon AI’s $500 million debt facility and $100 million capital raise aren’t pocket change for a neocloud.​

I’m bullish on what this means for enterprise AI adoption in Australia. The combination of 1,024 Blackwell Ultra GPUs with Cisco’s proven security portfolio gives regulated industries — banking, defence, healthcare — a credible on-shore alternative to shipping sensitive data through hyperscaler pipelines. That’s the unlock. The sandbox environment for proof-of-concept testing is a smart play too; it lowers the barrier for organisations that are AI-curious but risk-averse.​

My one reservation? Execution risk. Sharon AI is still a relatively young neocloud with ambitious expansion targets. Scaling from 1,024 to 20,000+ GPUs will test both their operational maturity and their financing model. But with Cisco and NVIDIA as partners — and the Australian government actively incentivising sovereign AI buildout through the National AI Plan — the tailwinds are strong.​

Overall, I’d rate this as a bullish development for Australia’s sovereign AI story. It is a strong signal that enterprise-grade, on-shore AI infrastructure is no longer a future promise — it’s being delivered today.

Add Sci-Tech Today as a Preferred Source on Google for instant updates!
google-preferred-source-badge
Joseph D'Souza
(Founder)
Joseph D'Souza founded Sci-Tech Today as a personal passion project to share statistics, expert analysis, product reviews, and experiences with tech gadgets. Over time, it evolved into a full-scale tech blog specializing in core science and technology. Founded in 2004 by Joseph D’Souza, Sci-Tech Today has become a leading voice in the realms of science and technology. This platform is dedicated to delivering in-depth, well-researched statistics, facts, charts, and graphs that industry experts rigorously verify. The aim is to illuminate the complexities of technological innovations and scientific discoveries through clear and comprehensive information.