Australia’s push to attract artificial intelligence investment is facing questions about jobs, energy and water use, public safety and whether communities will share in the benefits, as a federal parliamentary committee prepares recommendations on how the technology should be regulated.
The joint select committee on artificial intelligence is due to report by the end of November. The Albanese government has said it plans to introduce legislation early next year covering standards for large data centre operators, intellectual property and artists’ rights, as well as responsible AI use.
Labor and Coalition members have already issued recommendations that would require AI companies to undergo safety checks, reserve computing capacity for Australian use and provide security agencies with access to advanced AI models. Committee deputy chair and Liberal senator Jonno Duniam has argued that moving too slowly could leave Australia unable to capture investment and strategic benefits.
The potential investment is substantial. Treasurer Jim Chalmers has described Australia as one of the world’s top 10 markets for data centre capacity. The ABC reported that about $150 billion in AI infrastructure is slated for construction in Australia over the next five years. Treasury’s submission to the inquiry says global investment by large-scale cloud computing providers is expected to exceed $1 trillion this year.
However, the inquiry has received more than 500 submissions, many raising concerns about the effects of AI and the infrastructure supporting it. These include the demands data centres place on electricity, water, construction capacity and local communities, as well as how the economic gains will be distributed between Australians and foreign-owned technology companies.
Economist Richard Denniss told the committee that governments should examine whether spending on AI systems could reduce spending on workers or require larger public budgets. He also questioned whether the business case for some AI companies assumes large-scale job losses. These are his assessments, rather than established forecasts of the effect AI will have on Australian employment.
Senator David Pocock has criticised the government’s approach, accusing it of being “starstruck” by major United States AI companies, including OpenAI and Anthropic. He has proposed a 5 per cent tax on data centre revenue and called for clearer answers about how facilities will benefit nearby communities and what protections will apply.
Treasury’s submission also identifies risks from the technology. It warns that AI could increase the scale and sophistication of information manipulation and radicalisation, lower barriers to harmful biological and chemical activity, and make some criminal planning easier. It also notes that women and girls face disproportionate harm from AI-enabled abuse, including deepfakes and targeted harassment.
Supporters of faster adoption argue that AI investment could bring capital, infrastructure and strategic capability to Australia. Critics say the speed of policy development risks leaving the public with too little input and insufficient safeguards. The committee’s report, due by the end of November, is expected to inform the government’s planned legislation early next year; the final rules and their practical effects have not yet been settled.
