Australian government establishes committee to oversee agency AI use

The Australian Government is bringing together experts from across government to form an AI Review Committee, aiming to support the public service to maximise the opportunities of AI, while responsibly managing potential risks and harms.

The Committee will provide expert, albeit non-binding, advice to government agencies on new AI use cases with a residual high-risk rating; new AI use cases an agency identifies as highly sensitive, novel or complex before they are deployed; and serious AI incidents after an agency has completed its own response and remediation activities.

The Committee supports the government's policy for the responsible use of AI in government, rolled out at the end of last year, by strengthening oversight for AI use that could materially affect people, services or public trust.

The Committee is comprised of six experts from across government.

These include Chris Fechner, CEO of the Digital Transformation Agency; Dr Kate Conroy, GM of the AI Safety Institute, Department of Industry, Science and Resources; and Carly Kind, Privacy Commissioner, Office of the Australian Information Commissioner.

It also includes Lieutenant General Michelle McGuinness, National Cyber Security Coordinator, National Office of Cyber Security; Jo Talbot, Acting Deputy Commissioner, Workplace Relations, Integrity and Enabling Services, Australian Public Service Commission; and Andrew Watson, Deputy Commissioner and Chief Data Officer, Australian Taxation Office.

“The Committee is not designed to replace agency accountability or duplicate existing governance processes,” said Lucy Poole, deputy CEO for Strategy, Planning and Performance.

“Agencies remain responsible for identifying, assessing, managing and monitoring the risks of their AI use cases."

Poole said that the committee is setting a clear, consistent approach to how government uses AI in higher-risk settings.

“This Committee brings the right expertise together to support agencies to move with confidence – strengthening safeguards, improving transparency and ensuring AI brings real benefits to Australians while maintaining trust," she said.

The Committee terms of reference will be published after the first meeting is held.

Last year, the Australian Government introduced an AI Plan for the Australian Public Service, aiming to improve government service delivery, policy outcomes, efficiency, and productivity, through substantially increasing the use of AI in government.

In January of this year, a new AI Assessment Framework was introduced by the NSW Government, intending to give state agencies clearer guidance, stronger safeguards and more efficient assessments of AI systems.