Taking another step towards regulating the development and use of artificial intelligence (AI) in Australia, the Australian Government through the federal Department of Industry, Science and Resources (Department) has released a discussion paper entitled Safe and responsible AI in Australia (Discussion Paper).

The department is seeking system-wide feedback on actions that can be taken across the economy on AI regulation and governance. The paper has been released alongside the recent rapid response information report on Generative AI delivered by the National Science and Technology Council.

What is the consultation about?

The consultation process is one of a series of Government initiatives in the AI space that aim to increase community trust and confidence in the development and use of AI, to encourage and support responsible AI practices in Australia and to develop forward-looking, risk-based approaches to AI development and deployment.

Recognising public trust is a key driver for widespread acceptance of AI technologies and systems, and that building public trust and confidence will involve considering whether further regulatory and governance mechanisms are required to mitigate emerging risks associated with the use of AI, the Discussion Paper seeks to identify potential gaps in the existing domestic governance landscape and any possible additional AI governance mechanisms to support the development and adoption of AI.

To help readers navigating the complexity associated with AI governance, the Discussion Paper:

• notes the significant risks posed by the increased application of AI, including algorithmic bias, privacy concerns associated with access to and application of large data sets, transparency in AI as well as potential competition issues posed by organisations that own large and rich datasets;
• outlines the existing domestic mechanisms that currently govern the potential risks of AI, including the data protection and privacy law, the consumer and competition law, the copyright law and the criminal law, as well as sector specific regulations;
• examines approaches taken by other nations on the governance of AI. It notes countries like Singapore favour voluntary approaches while New Zealand and Canada have adopted a classification approach that divides AI into different risk categories with specific requirements applied to different government agencies;
• provides an overview of current Australian Government initiatives/works relevant to AI; and
• proposes a possible draft risk management approach for managing AI risks that caters to the context-specific risks of AI with requirements that can change depending on how the AI is deployed.

How to have your say

The Discussion Paper lists 20 specific questions to which the Government is seeking responses.

Readers are encouraged to answer some or all of the questions in the Department’s online survey or upload a separate submission.

The consultation will close on 26 July 2023.