Submission 211
Australian Catholic University (ACU)
Submitter group: Universities.
An AI model identified the submitter group and wrote the summary, key points and recommendation summaries. Counts do not measure importance or merit.
Summary
Read the original submission for the full text.
Draws on an ACU-led Delphi study of 95 digital health experts to argue that AI benefits depend on investment in trusted data, interoperability, governance, sector expertise and workforce capability alongside models and compute. Recommends sector-based ecosystems, health data stewardship, lifelong AI education, Commonwealth health AI accountability and child safeguards; notes the study did not focus on children.
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.
Argues that full AI benefits require investment in trusted data, interoperable infrastructure, governance, expertise and workforce, not only models and compute.
Australia will not achieve the full benefits of AI unless investment in models and compute is accompanied by investment in trusted data, interoperable infrastructure, governance, sector expertise, education and workforce capability.
Says the Delphi study identified fragmented systems, inconsistent standards, limited interoperability and uneven governance as barriers to digital transformation that constrain AI quality, scalability and safety.
The Delphi study identifies fragmented information systems, inconsistent standards, limited interoperability and uneven governance as barriers to effective digital transformation. These conditions constrain the quality, scalability and safety of AI.
States that data sovereignty concerns authority over access, use, quality and standards, not only data location.
Data sovereignty is not only about location. It also concerns authority over access, use, quality, standards, accountability and the conditions under which data may be used to develop, validate and evaluate AI.
Recommends a lifelong AI education and workforce approach from foundational learning through to continuing professional development.
Develop a coordinated, lifelong AI education and workforce approach, from age- appropriate foundational learning through tertiary education and continuing professional development, combining AI and data literacy with sector knowledge, implementation, ethics and governance.
Argues that children are not simply smaller adult users, and that their developmental stage, evolving capacity, privacy interests and dependence on institutions require specific consideration.
Children are not simply smaller adult users: their developmental stage, evolving capacity, privacy interests and dependence on institutions require specific consideration.
Recommendations
Recommendations are shown beside the submission’s own words. Listed in the same order as on its page in the interactive guide.
Treat data governance, interoperability and digital infrastructure as foundational enablers in national AI policy and investment.
Treat data governance, interoperability and digital infrastructure as foundational enablers in national AI policy and investment.
Apply the Delphi framework as one input to a national, sector-based AI readiness assessment, beginning with data-intensive sectors such as healthcare.
Apply the Delphi framework as one input to a national, sector-based AI readiness assessment, beginning with data-intensive sectors such as healthcare.
Build sovereign AI capability in health through stewardship of Australian data, standards, clinical information models and domain knowledge, alongside models, compute and technical talent.
Build sovereign AI capability in health through stewardship of Australian data, standards, clinical information models and domain knowledge, alongside models, compute and technical talent.
Support collaborative sector-based AI ecosystems spanning the research-to-impact value chain and combining technical, clinical, educational, behavioural, ethical, professional and community expertise.
Support collaborative sector-based AI ecosystems spanning the full research-to-impact value chain and combining technical, clinical, educational, behavioural, ethical, professional and community expertise.
Develop lifelong AI education and workforce capability from age-appropriate foundational learning through tertiary education and continuing professional development, combining AI/data literacy, sector knowledge, implementation, ethics and governance.
Develop a coordinated, lifelong AI education and workforce approach, from age- appropriate foundational learning through tertiary education and continuing professional development, combining AI and data literacy with sector knowledge, implementation, ethics and governance.
Require AI adoption in Commonwealth health-related services and administration to include clear accountability, transparent governance, appropriate human oversight and independent evaluation of safety, service quality, equity and workforce effects.
Require AI adoption in Commonwealth health-related services and administration to include clear accountability, transparent governance, appropriate human oversight and independent evaluation of safety, service quality, equity and workforce effects.
Include safeguards for children and vulnerable Australians: child-centred design, safeguarding, age-appropriate transparency, meaningful participation, developmental/wellbeing evaluation, child-centred research, professional practice and affected-community participation.
AI policy should therefore include child-centred design, safeguarding, age-appropriate transparency, meaningful participation and evaluation of developmental and wellbeing effects. Include specific safeguards for children and vulnerable Australians, supported by child-centred research, professional practice and participation by affected communities.
Assess whether current national digital-health initiatives collectively provide coherent foundations for next-generation systems and AI.
Australia is already investing in national digital health strategies, My Health Record modernisation, interoperability, standards and clinical information modelling. The study supports this direction while identifying the need to assess whether current initiatives collectively provide coherent national foundations for next- generation systems and AI.
Assess productivity across the full value chain, including safer information flows, less duplication, continuity of care, workforce experience, innovation adoptability and effective public investment; test whether expected benefits materialise.
Productivity should be assessed across the full value chain, including safer information flows, reduced duplication, continuity of care, workforce experience, adoptability of innovation and effective use of public investment. Evaluation should test whether anticipated benefits are realised in practice.
Recognise economic, social, cultural, community and policy impacts; assess AI benefits through productivity, services, living standards and the interests of metropolitan, suburban, regional and remote communities.
AI policy should recognise economic, social, cultural, community and policy impact. Given AI’s potential to reshape Australia’s economic, educational and social life, its benefits should be assessed by whether it improves productivity, service delivery and living standards while working in the interests of Australians across metropolitan, suburban, regional and remote communities.
Evaluate public-sector and community impacts robustly without reducing them to commercialisation, patents or investment measures alone.
In health, education and human services, value is often generated through collaboration with public-sector and for-purpose partners, translation into professional practice, improved service delivery and better outcomes for communities. These forms of impact should be evaluated robustly without being reduced to commercialisation, patents or investment measures alone.
Preserve interdisciplinary, applied, professional, mission-led and community-engaged capability in policy settings rather than privileging easily measured technical or commercial outputs.
Policy settings should preserve interdisciplinary, applied, professional, mission-led and community-engaged capability rather than privilege easily measured technical or commercial outputs.
Co-design sector-based AI initiatives with government, industry, professionals, service providers and affected communities, incorporating implementation conditions, lived experience and public value from the outset.
Sector-based AI initiatives should be co-designed with government, industry, professionals, service providers and affected communities so that technical development is informed by implementation conditions, lived experience and public value from the outset.
Support research-informed teaching, curriculum renewal and preparation of future professionals as part of AI education.
It should support age-appropriate AI literacy, research-informed teaching, curriculum renewal, preparation of future professionals and continuing development for the existing workforce.
Equip workers and leaders to navigate sectoral economic, legal, social and ethical implications; in regulated and care-based professions, include recognising limitations, maintaining human accountability and knowing when professional judgement must prevail.
This requires more than technical proficiency: workers and leaders must be equipped to navigate the economic, legal, social and ethical implications of AI within their sectors. In regulated and care- based professions, capability must include recognising limitations, maintaining human accountability and knowing when professional judgement must prevail.
Support domestic research training, emerging interdisciplinary capability and collaboration across national and global networks through workforce policy.
Workforce policy should also support domestic research training, emerging interdisciplinary capability and collaboration across national and global networks.
Require data-driven or automated assessments to be transparent, auditable and subject to appropriate expert judgement, with identifiable and enforceable responsibility.
A human-centred approach requires any data-driven or automated assessment to be transparent, auditable and subject to appropriate expert judgement, with responsibility remaining identifiable and enforceable.