AI Risk & Cybersecurity

Acronis researchers identified 575 malicious skills across the OpenClaw ecosystem. Snyk found roughly one in five agent skills is outright malicious. The AI distribution layer has become a primary supply chain attack surface, and Australian regulated industries have direct CPS 230 and CPS 234 exposure.

Frontier LLMs have collapsed the bug-to-exploit window from five months to ten hours, and agents have become a brand new attack surface. Here is what Australian regulated businesses need to do this quarter, and how it maps to the Essential Eight, CPS 234 and the Privacy Act reforms.

When an AI system is involved in a security incident, standard incident response procedures are not always adequate. Here is how Australian businesses should approach AI-specific security incidents, from detection through to recovery and reporting.

The Australian Cyber Security Centre's Essential Eight is the baseline cybersecurity standard for Australian organisations. AI changes how several of the eight controls need to be interpreted and implemented. Here is what you need to know.

Most AI security problems stem from decisions made early in implementation. Building security into AI systems from the start is far cheaper and more effective than retrofitting controls after something goes wrong.

AI has transformed phishing from a numbers game into a precision weapon. Here is how to recognise the new generation of AI-powered social engineering attacks and what Australian businesses can do to reduce their exposure.

Attackers are using AI to hit faster, at scale, and with far less human effort than before. Understanding the specific tactics they are deploying is the first step to building defences that actually work.

AI has changed cybersecurity on both sides of the fence. Australian businesses are facing a new class of threats that existing defences were not built for. Here is what the landscape looks like now and what it means for your organisation.