Ms Dianne Brown1, Komathy Padmanabhan, Jacky Cho, Sasenka Abeysooriya
1Monash University, Australia
Biography:
To be confirmed
Abstract:
Situation
AI presents a transformative opportunity for research. In the global race to ride the AI wave, governments, industry and research institutions are making significant investments in AI infrastructure and talent development. However, this momentum often overlooks a critical enabler of AI success: data. Without trustworthy, high-quality, and ethically managed research data, even the most advanced AI systems risk producing misleading or unethical results, undermining the integrity of scientific discovery.
Task/ Action
Australia must ensure its institutional and national research data assets are AI-ready. Opportunities exist in:
Creation of a repository of curated training and validation datasets, underpinned by well-maintained data catalogues
Creating data quality standards and practices to ensure AI systems are trained on research data that produces valid, meaningful, and reproducible insights.
Defining researcher’s role (and training them) as data stewards who ensure ethically curated, well-documented, and fit for training AI-ready data that also uphold other data principles such as CARE Creation of storage and infrastructure that considers long-term data access, versioning, and scalability.
Managing AI-specific research data risks—such as bias, misuse, security, IP and privacy. This will be central in the responsible and ethical use of AI in research and maintaining research’s ongoing social license.
Result
A chance for a discussion around the evolution of research data governance, Indigenous data sovereignty, research ethics and integrity processes to account for these challenges and the new requirements for provenance, documentation of data use, persistent identifiers, consent models and other accountability measures to support researchers navigate this changing landscape.