. Showing 1 to 7 out of 7 results.
| Date Sort ascending | Title | Summary | Publisher |
|---|---|---|---|
| 25 Nov 2025 | Second key update: technical safeguards and risk management | This report looks at developments in technical approaches to general-purpose AI risk management since the ‘International AI safety report’ 2025. | AI Safety Institute |
| 15 Oct 2025 | First key update: capabilities and risk implications | This report covers the major breakthroughs in AI capabilities since the ‘International AI safety report’ 2025. | AI Safety Institute |
| 29 Jul 2025 | Risk analysis techniques for governed LLM-based multiagent systems | This report from the Gradient Institute looks at the risks of multiple AI agents interacting with each other. | AI Safety Institute |
| 17 Jul 2025 | Advancing methods for agentic evaluations across domains | Australia contributed to this joint testing of AI ‘agents’ (AI systems that can plan steps and use tools to do tasks) to improve how countries test these systems for risks like data leaks, fraud and cyber threats. | AI Safety Institute |
| 10 Feb 2025 | Improving international testing of foundation AI models | Australia contributed to this joint testing of AI models to more accurately measure AI risks and to strengthen our collaboration with network partners. | AI Safety Institute |
| 3 Feb 2025 | International AI safety report 2026 | Australia’s AI experts contributed to the second comprehensive review of scientific research on the capabilities and risks of general-purpose AI systems. | AI Safety Institute |
| 29 Jan 2025 | International AI safety report 2025 | Australia contributed to the first comprehensive review of scientific research on the capabilities and risks of general-purpose AI systems. | AI Safety Institute |