A new parliamentary inquiry is examining the opportunities and challenges presented by artificial intelligence and its potential impact on Australian society.

The Joint Select Committee on Artificial Intelligence was appointed by the Australian Parliament on 20 August 2026 to examine a range of issues, including AI’s impact on productivity, the economy, the workforce, public services, national security and Australian society. 

Researchers from the ARC Centre of Excellence for Automated Decision-Making and Society (ADM+S) have contributed to the inquiry, with a submission drawing on research across computer science, law, social sciences, communications and the humanities.

The submission addresses ten key areas of inquiry from AI and productivity and digital inclusion to data sovereignty, AI in the public sector, adequacy of law, regulatory frameworks for data centres, risks and harms and equipping Australia’s regulators.

Across these areas, researchers highlight the need for Australia to support AI innovation while ensuring systems can be understood, governed and held accountable. The submission calls for stronger research and Australian capability, improved digital inclusion, clearer legal and regulatory frameworks, and greater opportunities for independent experts to monitor and scrutinise AI systems.

The submission highlights the difficulty of assessing claims about AI’s productivity benefits.

ADM+S research with senior members of the Victorian public service suggests that short- to medium-term productivity gains may be overestimated, while longer-term benefits from deeper AI integration require significant organisational preparation, time and investment.

The researchers argue that evaluating productivity should be part of AI adoption from the outset, rather than something considered after systems have been introduced.

They also call for workers to have a role in evaluating AI, noting that workers can help identify both benefits and less visible costs, including increased error rates or impacts that may fall disproportionately on vulnerable groups.

The submission also warns that the benefits of AI will not be shared equally unless Australia addresses existing digital inequalities.

Research from ADM+S shows that 1 in 5 Australians experience high levels of digital exclusion, with significantly higher levels among First Nations Australians and people living in remote and very remote areas.

The researchers argue that digital access should be recognised and included as an essential service when planning new developments and that AI literacy should be part of a coordinated national approach to ensuring Australians can participate in and benefit from AI.

Improved digital and AI literacy could also help reduce AI harms such as misinformation and scams, while inclusive technology design can deliver benefits across the broader population.

ADM+S researchers argue that sovereignty should not be understood simply as Australia owning every part of the AI technology stack. Instead, it should include the ability to understand, govern, build and control AI systems used in Australia.

This means maintaining choices about which technologies and providers Australia relies on, reducing the risk of being locked into particular commercial systems, and developing local technical expertise and infrastructure.

The researchers call for stronger legal protections around the use of AI and automated decision-making by government.

They argue that voluntary standards and government policies that are not legally enforceable are not a substitute for legislation, particularly where automated systems can affect people’s rights and access to services.

ADM+S has been involved in work with Australian governments and regulators on the governance of automated decision-making, including following the recommendations of the Royal Commission into the Robodebt Scheme.