Submission 210
Australian Academy of the Humanities
Submitter group: Research institutes.
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.
Argues that Australia will not secure the benefits of AI through technical capacity alone and that humanities, arts and social science expertise should be part of national AI capability. Recommends a National AI Research Plan, protection of knowledge and data sovereignty, and a broader, skills-based conception of AI capability, with case studies.
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 australia will not secure the benefits of AI through technical capacity alone and needs wider research, data, institutional and workforce capabilities.
Australia will not secure the benefits of AI through technical capacity alone. It needs the research, knowledge, data, institutional and workforce capabilities required to design, adapt, evaluate and govern AI in Australian contexts.
States that a public policy mechanism to draw on humanities and arts expertise for the National AI Plan's goals has not been established.
The goals of the National AI Plan are socially- situated and community-focused. Achieving them requires foundational engagement with the insights of humanities and arts capability and research: language and linguistics, history, philosophy, ethics, literary studies, cultural studies, media and communication, creative arts. A public policy mechanism to draw on this expertise has not been established.
Argues that sovereign AI capability depends partly on access to high-quality Australian humanities, cultural, linguistic, archival and First Nations data.
Australia’s ability to develop useful, sovereign and competitive AI technologies and applications will depend in part on access to high-quality Australian data. This includes humanities, cultural, linguistic, archival and First Nations data that can support domain- specific AI applications.
Recommends strong data governance and access protocols, with First Nations leadership essential for AI involving First Nations knowledges.
strong data governance and access protocols, with First Nations’ leadership essential for any consideration of AI with regard to First Nations knowledges
Argues for moving from occupation-based AI expertise to national combinations of knowledge, skills and judgement to create, use, evaluate and govern AI.
It invites us to change our thinking from a narrow occupation-based model of AI expertise (who are the technical AI specialists?) to a national capability-based model (what combinations of knowledge, skills and judgement does Australia need to create, use, evaluate and govern AI across the economy?).
Argues that cultural and social dimensions must be embedded at all layers of AI systems and throughout the AI lifecycle.
Cultural and social dimensions must be embedded at all layers of AI systems and throughout the AI lifecycle: formulating problems, designing systems and data selection, translating into applications, and their evaluation and deployment.
Recommendations
Recommendations are shown beside the submission’s own words. Listed in the same order as on its page in the interactive guide.
Develop a National AI Research Plan to build and steward sovereign national research capability to design, develop, adapt, evaluate and govern AI.
Develop a National AI Research Plan: To build and steward sovereign national research capability to design, develop, adapt, evaluate and govern AI.
Coordinate the cross-portfolio National AI Research Plan through PM&C, involving Industry, Science and Resources, Education and other agencies.
This is necessarily a cross-portfolio agenda involving the Department of Industry, Science and Resources, Department of Education, and other agencies, so PM&C is a natural coordination point.
In the National AI Research Plan, secure buy-in and insights from across the full spectrum of research and earn the confidence of industry and the community.
Buy-in and insights from across the full spectrum of research (i.e. humanities, arts, Indigenous research, sciences, social sciences, health, medical sciences, technological sciences) and it needs to earn the confidence of industry (broadly defined) and the community.
In the Plan, back and build humanities-led R&D and scale it nationally (including through the ARC's new National Competitive Grants Program from 2027) and internationally through Horizon Europe associate membership.
To back and build humanities-led R&D, which has already delivered significant advances and community benefits, but is not yet widely recognised or systematically supported. There are major opportunities for the Australian Government to engage with the outcomes of these work programs, and to scale nationally (including through the ARC’s new National Competitive Grants Program from 2027), and internationally through Australia’s new status as an associate member of Horizon Europe (from January 2027).
Revisit the place of AI in Australia's Science and Research Priorities to put people and communities at the centre of AI research.
To revisit the place of AI in Australia’s Science and Research Priorities (NSRPs). AI is explicitly mentioned in two contexts only: ‘Transitioning to a net zero future’, and ‘Building a secure and resilient nation’. We need stronger signals to drive programs that put people and communities at the centre of Australia’s AI R&D.
In the Plan, drive a 'national capability' approach to AI with clear roles for the Australian Tertiary Education Commission, drawing on Jobs and Skills Australia's 'skills first' approach.
To drive a ‘national capability’ approach to AI with clear roles for the Australian Tertiary Education Commission (ATEC) and drawing on Jobs and Skills Australia’s (JSA) ‘skills first’ approach (see below).
In the Plan, recognise innovation capability where technologies are adapted and deployed, and establish durable mechanisms connecting universities, government and industry for interdisciplinary AI research and translation (in the Government's response to the Ambitious Australia report).
Recognise innovation capability not only where new technologies are invented, but where technologies are adapted, interpreted and responsibly deployed in complex human environments. The Government’s response to the Ambitious Australia (2026) report needs to establish durable mechanisms connecting universities, government and industry to support interdisciplinary AI research and translation.
In the Plan, factor in and plan for Australia's participation in Horizon Europe.
Factor in and plan for Australia’s participation in Horizon Europe. Europe’s agenda for AI changes the ‘game’ for Australian humanities research on and with AI.
Provide long-term support for digitisation, curation and infrastructure for high-value Australian cultural and research collections.
Investment in Australian AI data capability includes the need for: • long-term support for digitisation, curation and infrastructure for high-value Australian cultural and research collections;
Support domain-specific datasets and models developed with the relevant research and knowledge communities.
domain-specific datasets and models developed with the relevant research and knowledge communities;
Establish strong data governance and access protocols, with First Nations' leadership essential for any consideration of AI regarding First Nations knowledges.
strong data governance and access protocols, with First Nations’ leadership essential for any consideration of AI with regard to First Nations knowledges; and
Fund research and infrastructure programs that bring humanities expertise into the development of Australian domain-specific AI.
research and infrastructure programs that bring humanities expertise into the development of Australian domain-specific AI.
Adopt a broader conception of AI capability to respond to AI's impacts on work, its ethical implications and barriers to its productive adoption.
A broader conception of AI capability is needed to respond to AI’s impacts on work, its ethical implications and removing barriers to its productive adoption. Jobs and Skills Australia (JSA) ‘skills first’ approach has potentially profound implications for how Australia should define AI capability.
Develop distinct educational pathways for applied AI professionals, technical specialists, citizen AI literacy, strategic leadership and governance expertise in law, ethics, rights, safety, trust and risk.
Different capabilities will require different educational pathways, such as: • Applied AI capability: Professionals need to know how to integrate AI safely, effectively and responsibly into their particular field. • Technical AI capability: Specialists need deep technical expertise; all citizens need an uplift in AI literacy. • Strategic AI capability: Leaders need to understand how AI changes organisations, industries, institutions and society. • Governance capability: Professionals and policymakers need access to specialist expertise in law and regulation, ethics and rights, security and safety, public trust, and risk.
Embed cultural and social dimensions across the AI lifecycle and build workforce capability connecting humanities, arts and social sciences with technical understanding.
Cultural and social dimensions must be embedded at all layers of AI systems and throughout the AI lifecycle: formulating problems, designing systems and data selection, translating into applications, and their evaluation and deployment. Achieving this will require a skilled Australian workforce able to connect humanities, arts and social science expertise with a strong understanding of the technological dimensions of AI systems.