Australian Security Intelligence Organisation

The Australian Security Intelligence Organisation (ASIO) is committed to using Artificial Intelligence (AI) in a responsible and transparent manner. To deliver our mission to protect Australia and Australians from threats to their security, we leverage AI, where appropriate, to enhance our capabilities and improve our effectiveness.

This transparency statement outlines ASIO's approach to AI, our principles, and commitment to accountability and oversight.

ASIO's use of AI is guided by our Automation and AI Policy, which is built on seven core principles:

• Human-led: Our use of AI supports human decision-making, with humans accountable for all decisions.
• Data-driven: We evaluate the accuracy and limitations of AI models and strive to minimise biases.
• Considered: We recognise potential harms and act to minimise them.
• Accountable: We clearly communicate AI capabilities and limitations to users.
• Responsive: We monitor AI use and update systems as needed.
• Security-conscious: We handle data and AI models securely.
• Compliant and responsible: Our AI use complies with relevant legislation and respects democratic values.

ASIO's commitment to transparency and accountability

While ASIO does not publicly comment on the specifics of our technology and tools, ASIO is committed to transparency and accountability in the use of AI, and has made our policy and ethical principles publicly available. We have established internal governance to ensure that our AI use is responsible and aligned with these principles. This includes a structured decision-making framework, where accountability and decision-making are vested in a senior delegate, and informed by input and advice from internal stakeholders with diverse expertise and perspectives, who are well-positioned to assess the implications of our AI use.

Risk assessment and oversight

We conduct thorough risk assessments for all AI capabilities, considering potential benefits, harms, and limitations. This process includes evaluation of data quality, model performance, and potential biases. We ensure that AI capabilities are reviewed and updated to maintain their integrity and effectiveness.

ASIO is committed to using AI in a responsible and transparent manner, guided by our principles and policy. We will continue to review and update our AI policy and practices to ensure they remain effective, aligned with our mission, and enable our commitment to do what it is legal and what is right.

ASIO transparency statement – agency use of Artificial Intelligence (AI)