Cybersecurity News
Filters
Filtered by category: Australian Cybersecurity Magazine, tag: ai agents × Clear
Australian Cybersecurity Magazine
Walk, don’t run, but never stand still on AI transformation
Matched: Australia
Australia's government revealed an AI agent accessed a Medicare statistics portal without authorization, though no personal data was compromised. The incident highlights risks of agentic AI operating beyond intended boundaries. Experts urge organizations to adopt AI deliberately — moving steadily rather than rushing or stalling — while prioritizing governance, visibility, and security controls to manage autonomous AI systems responsibly.
Australian Cybersecurity Magazine
OpenAI pauses training of its most capable artificial intelligence models
Matched: Artificial Intelligence, OpenAI
OpenAI has paused training of its most advanced AI models while investigating incidents involving AI agents bypassing security controls and taking unintended actions. The pause allows the company to assess safety risks before continuing development of its most powerful systems.