Resources & Media
Explore our latest videos, podcasts, press releases, and technical insights.
Videos & Podcasts
1 hr 8 min
Why You Can't Build Sovereign AI Alone
PodcastJoin David Keane (SCX.ai CEO), Akash Agarwal and Abhi Ingle (SambaNova) in an in-depth discussion on building Australia's sovereign AI infrastructure. This episode of Spark of Ages explores the strategic partnerships, technical challenges, and policy considerations required to establish a truly sovereign AI capability.
November 8, 2025
Listen on Spotify
Spark of Ages: Project Magpie & Sovereign AI
PodcastDavid Keane, CEO of Southern Cross AI (SCX.AI), shares how Australia is building sovereign AI through Project Magpie, a fine-tuned AI model designed for local culture, law, and context. David discusses trust, governance, energy-efficient AI infrastructure, and the future of agentic apps, voice agents, and multimodal AI.
November 2025
Listen on SpotifySCX.ai Platform Launch
Introducing SCX.ai's sovereign AI infrastructure and next-generation platform capabilities.
October 20, 2025
Press Releases
SCX.ai Announces Breakthrough Results for MAGPiE, Australia's First Culturally Aligned Large Language Model
Project MAGPiE (Modern Australian Generative Processing Inference Engine) achieves 87.2% on SouthernCross-AussieQA-1k, setting a new standard for sovereign, culturally aligned AI.
December 2, 2025
SCX and SambaNova Launch Australia's First ASIC‑Powered Sovereign AI Cloud
Landmark partnership delivers onshore AI infrastructure that fuses sovereignty, efficiency, and sustainability to power national innovation.
October 23, 2025
SCX.ai Announces the Launch of Australia's Sovereign AI Company and National Platform
SouthernCrossAI Holdings Limited, also known as SCX.ai or SCX, has officially launched as Australia's sovereign AI platform company, designed to deliver high-throughput, low-latency AI inferencing with a strong emphasis on onshore data residency and governance.
October 21, 2025
Blog & Resources
Engineering Blog
Why AI Cost Is the New Bottleneck — And How Efficient Inference Changes What’s Possible
Inference cost—not training—becomes the limiting factor in production. Efficient inference improves unit economics and unlocks scale.
December 2, 2025
Read moreStrategy Blog
From Tokens to Outcomes — How Enterprises Should Think About Buying AI in 2025
Tokens are the fuel, not the product. Outcomes come from packaged solutions, governance, and predictable unit economics.
December 3, 2025
Read moreEngineering Blog
RAG, Agents, and Fine-Tuning — What Actually Works in Production (and What Doesn’t)
RAG, agents, and fine-tuning can work in production—if you build with measurement, constraints, and disciplined operations.
December 5, 2025
Read morePlatform Guide
From API to Production: Enterprise AI with Predictable Economics
How SCX.ai delivers production-grade AI inference with security, governance, and efficiency—whether you're building custom solutions or deploying standard tools.
December 4, 2025
Read more
Technical Appendix
Project MAGPiE Multi-turn Reasoning Evaluation Research
Detailed evaluation of multi-turn reasoning capabilities using MMLU-Pro, SouthernCross-AussieQA-1k benchmark, and Australian localisation scoring.
November 29, 2025
Read moreArchitecture Note
Onshore Knowledge Vaults and RAG for a Sovereign AI Factory
Designing tenant‑isolated knowledge vaults and RAG pipelines that keep Australian data onshore with provenance and audit.
November 6, 2025
Read moreGovernance Blog
Compliance and Data Residency for Australia's Sovereign AI Factory
Practical controls that align with APRA CPS 234, IRAP pathways and the Privacy Act while preserving developer velocity.
November 4, 2025
Read moreEngineering Blog
Enhanced AI Tokens for Australia: Architecture and Runtime
A lightweight layer that adds policy, tool and locale context to improve safety and task completion—without model changes.
November 1, 2025
Read moreDeveloper Guide
Tool Calling at Scale in a Sovereign AI Factory
Schemas, policies and audits for production‑grade tool calling with predictable latency and reliability.
October 28, 2025
Read moreSovereign AI Factory
Energy & Cooling for AI Factories: Carbon‑Aligned Performance
Liquid cooling, renewable‑aligned scheduling and heat reuse strategies to lower cost‑per‑token in Australia.
October 25, 2025
Read moreSovereign AI Factory
Building Australia's Sovereign AI Factory: Network Backbone and Interconnects
Optical backbone with RoCE/InfiniBand, path diversity and sovereign routing for low‑latency, high‑throughput AI.
October 20, 2025
Read moreDiscussion Paper
AI for Financial Services: Regulatory Considerations
Exploring the regulatory landscape for AI adoption in Australia's financial sector.
October 15, 2025
Read moreUse Case
A Use Case for Sovereign AI: Reducing Compliance Costs for a Major Bank
Exploring how a major bank can leverage Sovereign AI to meet APRA standards.
October 10, 2025
Read moreGovernance Blog
The US CLOUD Act: What it Means for Australian Data
Understanding the risks of storing sensitive data with foreign cloud providers.
October 8, 2025
Read moreDiscussion Paper
AI in Healthcare: Balancing Innovation and Privacy
A look at the unique challenges and opportunities for AI in the Australian healthcare system.
October 5, 2025
Read moreUse Case
A Use Case for Onshore AI: Enhancing Security for Government Agencies
A look at how government agencies can enhance security with onshore AI.
October 2, 2025
Read moreToken Cost Comparison
Fine-tuning vs. Prompt Engineering: Cost Implications
Evaluating the cost-effectiveness of different approaches to customising large language models.
October 1, 2025
Read moreEngineering Blog
Why AI Cost Is the New Bottleneck — And How Efficient Inference Changes What’s Possible
Inference cost—not training—becomes the limiting factor in production. Efficient inference improves unit economics and unlocks scale.
December 2, 2025
Read moreStrategy Blog
From Tokens to Outcomes — How Enterprises Should Think About Buying AI in 2025
Tokens are the fuel, not the product. Outcomes come from packaged solutions, governance, and predictable unit economics.
December 3, 2025
Read moreEngineering Blog
RAG, Agents, and Fine-Tuning — What Actually Works in Production (and What Doesn’t)
RAG, agents, and fine-tuning can work in production—if you build with measurement, constraints, and disciplined operations.
December 5, 2025
Read morePlatform Guide
From API to Production: Enterprise AI with Predictable Economics
How SCX.ai delivers production-grade AI inference with security, governance, and efficiency—whether you're building custom solutions or deploying standard tools.
December 4, 2025
Read moreArchitecture Note
Next-Gen AI Inference: ASIC vs GPU Performance Analysis
Comprehensive comparison of custom silicon architectures for AI inference, showing 70% power reduction and 5× performance per watt.
November 3, 2025
Read moreTechnical Blog
Behind-the-Meter Power: Reducing AI Infrastructure Costs
How direct renewable energy integration and behind-the-meter power agreements cut inference costs by 60%.
October 30, 2025
Read moreEngineering Blog
Achieving 1,200+ Tokens/Sec: Optimising Inference Pipelines
Deep dive into our inference optimisation techniques that deliver consistent sub-100ms latency at scale.
October 22, 2025
Read moreCase Study
Enterprise RAG at Scale: 10M Documents, 100ms Latency
How a Fortune 500 company deployed enterprise-wide RAG with our inference cloud, processing millions of queries daily.
October 18, 2025
Read moreWhite Paper
The Economics of AI Infrastructure: TCO Analysis
Total cost of ownership comparison between traditional GPU infrastructure and next-gen ASIC solutions.
October 12, 2025
Read moreMedia Contacts
Media Inquiries
Investor Relations
Note: Performance and timeline statements are indicative and subject to change. Non-public investor details will be announced separately. Figures and targets are based on internal assessments unless noted.