AI Inquiry Atlas

An unofficial, independent guide to the public records of the Joint Select Committee on Artificial Intelligence. Not connected with the Parliament of Australia or the committee. Its makers also made a submission (no. 481; see About). Summaries and scores are written by AI and can be wrong: check the original before relying on them.

Submission 231

CPA Australia

Submitter group: Industry bodies.

An AI model identified the submitter group and wrote the summary, key points and recommendation summaries. Counts do not measure importance or merit.

Open in the interactive guide

Summary

Read the original submission for the full text.

Argues that AI adoption rates are a poor measure of success and that the challenge is integrating AI into business processes, especially for SMEs. Presents survey findings and makes five recommendations on capability-focused support, SME programs, skills and judgement, government AI accountability, and proportionate risk-based regulation.

Key points

Quotes are the submitter’s own words. Text taken from documents, including scanned pages read by text-recognition software, can differ from the PDF. Check the original. Listed in the same order as on the submission’s page in the interactive guide.

  1. Describes AI adoption as tool uptake and integration as embedding AI into business processes, systems and decision-making.

    While adoption measures the uptake of AI tools, integration refers to embedding AI into business processes, systems and decision-making.

  2. States that its 2026 survey found widespread business use of AI but limited deep integration.

    CPA Australia's 2026 Business Technology Survey found that 88 per cent of surveyed Australian businesses used AI in some form in the past 12 months. Despite this, only 10 per cent reported that AI is widely used across the business and is deeply embedded.

  3. States that its research suggests the primary barrier for small businesses is implementation capability rather than awareness.

    CPA Australia research suggests that the primary barrier to AI adoption is not awareness of the technology, but the capability required to implement it effectively and realise business benefits.

  4. Says traditional learning through routine work may become less effective as AI automates entry-level tasks.

    Traditional models that rely heavily on learning through routine work may become less effective as AI automates many entry-level tasks.

  5. Argues that responsibility for government decisions should remain with identifiable public officials rather than technology systems.

    Responsibility for decisions should remain with identifiable public officials rather than technology systems.

  6. Argues that regulation should be proportionate to risk, with stronger rules possible for higher-risk applications and lighter burden for lower-risk workplace tools.

    Regulatory approaches should be proportionate to the risks. For higher-risk AI applications, stronger regulation may be appropriate. This could include clear accountability, documentation, testing, monitoring and human oversight arrangements proportionate to the potential impact of the AI system. Lower-risk applications, including many workplace productivity tools, should not be subject to the same level of regulatory burden.

Recommendations

Recommendations are shown beside the submission’s own words. Listed in the same order as on its page in the interactive guide.

  1. Support organisational change, business process redesign, workforce development and leadership capability to turn AI adoption into measurable productivity and competitiveness gains.

    This requires support for organisational change, business process redesign, workforce capability development and leadership capability so that AI adoption translates into measurable productivity, competitiveness and economic gains.

  2. Focus government policies and support programs on business and workforce capability to integrate AI effectively, rather than adoption alone.

    Government policies and support programs should focus on building business and workforce capability to effectively integrate AI rather than adoption alone.

  3. Make AI policy support and regulation proportionate to business size, complexity and risk profile.

    AI policy support measures and regulation should therefore be proportionate to the size, complexity and risk profile of the business.

  4. Federal and state governments to significantly expand current digital support programs and incentives for SMEs, prioritising practical AI implementation support (advisory services, workforce training, responsible AI guidance) over awareness-building or adoption targets.

    Federal and state governments should significantly expand the scale and reach of their current digital support programs and incentives for SMEs. These programs should prioritise practical AI implementation support, including access to trusted advisory services, workforce training and responsible AI guidance, rather than focusing primarily on awareness-building or adoption targets.

  5. Support AI literacy while developing alternative pathways for professional judgement, critical thinking and communication skills, including mentoring, simulations, work-integrated learning and case-based assessment.

    Workforce and education policies should support AI literacy while also developing alternative pathways for building professional judgement, critical thinking and communication skills, including mentoring, simulations, work-integrated learning and case-based assessment.

  6. Hold government AI use to at least the accountability, transparency and governance standards expected of businesses.

    Government should be held to at least the same standards of accountability, transparency and governance that it expects of businesses using AI.

  7. Keep responsibility for government decisions with identifiable public officials rather than technology systems.

    Responsibility for decisions should remain with identifiable public officials rather than technology systems.

  8. Ensure government AI-enabled decision-making is supported by transparency, accountability, meaningful human oversight, review mechanisms and the ability to challenge decisions where AI has materially influenced an outcome.

    Government AI frameworks should ensure AI-enabled decision-making is supported by transparency, accountability, meaningful human oversight, appropriate review mechanisms and the ability to challenge decisions where AI has materially influenced an outcome.

  9. Have future reforms provide certainty while avoiding unnecessary compliance costs.

    Future reforms should therefore seek to provide certainty while avoiding unnecessary compliance costs.

  10. Consider stronger regulation for higher-risk AI, with proportionate accountability, documentation, testing, monitoring and human oversight; avoid imposing the same burden on lower-risk applications, including many workplace productivity tools.

    Regulatory approaches should be proportionate to the risks. For higher-risk AI applications, stronger regulation may be appropriate. This could include clear accountability, documentation, testing, monitoring and human oversight arrangements proportionate to the potential impact of the AI system. Lower-risk applications, including many workplace productivity tools, should not be subject to the same level of regulatory burden.

  11. Adopt a proportionate, risk-based and technology-neutral approach to AI regulation that builds on existing frameworks wherever possible.

    Australia should adopt a proportionate, risk-based and technology-neutral approach to AI regulation that builds on existing frameworks wherever possible.

Read the original

Open in the interactive guide