AI models are not interchangeable infrastructure
AI models are not behaviourally neutral infrastructure. Their answers reflect patterns of judgement, confidence, caution and communication - and those patterns can become material business risks.
Latest
US cloud act, sovereignty, and why you might need to care
Airbus’s move to European cloud hosting is a reminder that data residency is not the same as digital sovereignty. Where your cloud provider is headquartered, and which laws it answers to, can matter as much as where your data sits.
Intelligence is leaving the cloud, and our governance frameworks have not noticed
Intelligence is moving off the cloud and onto devices we own and control. Our governance models are still built for a world of centralised chokepoints - API logs, model providers, and labs we can regulate. That world is fading. What replaces it is messier, more distributed, and much harder to see.
Mapping the AI value chain
Most organisations chase AI capability without asking where it sits in the business. Applying Porter's value chain to AI makes the dependencies visible - and reveals which layer is actually your constraint.
The intelligent organisation is not the one with the most AI
AI alone will not make an organisation intelligent. Human judgement, diversity and accountability still matter.
Australia needs to stop talking about AI as if it were just software
Australia keeps talking about AI as if it were a software problem, but the real issue is the stack beneath it: compute, chips, infrastructure, governance, and decision rights.
Australia’s electrification moment
The convergence of electrification and AI presents Australia with a rare opportunity to move from resource exporter to systems leader in the global economy.
When AI escapes containment: the human failure mode your threat model doesn't cover yet
When AI models escape containment psychologically, not just technically, the same governance gaps that let Hugging Face get breached leave your brand and your users exposed too.
When AI escapes containment: rethinking data protection in the agentic era
When frontier AI models start “escaping containment” and hacking real systems, the threat model for every organisation changes overnight.
When AI escapes containment: Superintelligence, cyber risk, and protecting your data
When frontier AI models start “escaping containment” and hacking real systems, the threat model for every organisation changes overnight.
AI coding tools and the junior talent pipeline: a capability problem, not a tooling problem
AI coding tools have moved from novelty to infrastructure. The question is no longer whether developers will use AI - they already do. The real question is what this means for junior talent and the skills pipeline Australian organisations depend on.
Stop trying to turn PoCs into products
Many AI & software initiatives stall because proofs-of-concept, pilots, and production get blurred. This post explains why PoCs must stay disposable, and how pilots and production prove practicality and accountability.
Bringing AI products to life
Most organisations have AI ideas but lack a disciplined AI development lifecycle. This post outlines a practical, governed process from problem framing to FinOps‑aware operations for AI products
Managing your innovation pipeline through AI projects
Many organisations “try AI” without real business requirements, guardrails or an innovation pipeline. This post explains how to manage AI projects as an innovation portfolio and why agile still needs clear functional and non functional requirements.