What Australia AI Safety Institute does
The Australian AI Safety Institute (AISI) is a government capability being established within Australia’s Department of Industry, Science and Resources (DISR) to help monitor, assess, and respond to emerging advanced-AI capabilities and the risks and harms they may pose over time. The institute’s stated remit includes analysing and testing new AI models and applications, supporting regulators and agencies with technical information on emerging AI risks and harms, and contributing to safe AI development, deployment, and international governance aligned with Australia’s interests. AISI also positions its work to complement Australia’s existing technology-neutral AI protections across privacy, consumer protection, online safety, workers’ rights, and anti-discrimination, and to support a coordinated approach across government portfolios. Operationally, AISI reports working with domestic and technical partners such as the Australian Signals Directorate and CSIRO, partnering closely with the National AI Centre (NAIC) to translate institute insights into practical guidance, and participating internationally through mechanisms including NAAIMES (the International Network for Advanced AI Measurement, Evaluation and Science, formerly called the International Network of AI Safety Institutes). AISI also publishes and contributes to safety-science outputs, including report updates and joint testing work aimed at improving measurement and evaluation of advanced AI systems, such as capabilities and risks in multi-agent or agentic settings. The institute is funded via a dedicated government appropriation measure described in Australia’s fiscal documents and has begun releasing public updates and publications in 2025–2026 as it ramps up its establishment and partnership activities.