AI transparency in practice

This report assesses how well Commonwealth government departments and agencies are complying with the requirement to publish an artificial intelligence (AI) transparency statement, a key aspect of the Policy for the responsible use of artificial intelligence AI in government. This requirement became mandatory for many Commonwealth government departments and agencies from February 2025.

The report concludes that AI transparency statements are not as easily found as they ought to be. Very few statements were accessible through ‘a link to the statement’ as recommended by the Digital Transformation Agency (DTA). The report identified 30 government entities potentially within the scope of the policy for which a statement could not be found. It also found that AI transparency statements vary significantly in the detail provided. Some are useful and informative, but others failed to comply with the requirements of the Standard for AI transparency statements. 

The report outlines recommendations that would improve the findability and informativeness of AI transparency statements.

Recommendations

• AI transparency statements should be published in one central location.
• The DTA should reconsider the entities subject to the Policy and explore mechanisms to ensure that the policy and requirements are complied with, including by considering what consequences flow from non-compliance.
• The Standard for AI transparency statements should be revised.