FAIR & AI Symposium @ TU Graz
Abstract
This record contains the presentations of the the FAIR Daniel Garijo argued that FAIR is a means to improve credit, reproducibility, and trustworthy AI, but quality gaps persist. Lightning talks covered HPC access, AI governance, and the evolving role of data stewards. Breakout groups explored how AI can aid FAIRification and how FAIR data strengthens AI. Participants agreed that standardized tools, certification, and automated support for repetitive tasks will be crucial. An overview of the event program can be found in the file: Overview. pdfThe report of the event can be found in the file: Report. pdf Keynote Presentations Jana Lasser: FAIR data vs. GDPR - a researchers perspective (see: FAIRᵥsGDPRLasser. txt) Daniel Garijo: Beyond FAIR for data: Quality in Heterogeneous Digital Objects (see: TowardsFAIRforAIGarijo) Lightning Talk Presentations Markus Stöhr: Infrastructure for HPC and AI workloads (see: InfrastructureforHPCₐndAIworkloadsᵢnAustriaₐndEuropeStöhr. pdf) Jeannette Gorzala: The Future runs on Trust Code (see: TheFutureᵣunsₒnTrustCodeGorzala. pdf) Emily Kate & Michael Feichtinger: A Long Way to FAIR and a Short Time to Get There: Small Steps Toward Big Promises (see: ALongWayₜoFAIRₐndShortTimeₜogetₜhereKateFeichtinger. pdf) Please check the license terms specified in the presentations for the reuse of the content. For more information, the event webpage is linked (see references).
Key Points
Objective
The aim was to investigate how FAIR research data principles and AI influence each other amidst emerging challenges.
Methods
- Gathered researchers, data stewards, and experts in policy and infrastructure.
- Conducted presentations, keynotes, and breakout groups discussing AI and FAIR practices.
- Highlighted privacy, transparency, and accountability issues in relation to data management.
Results
- AI tools can enhance FAIR practices through automated metadata and better data discoverability.
- Concerns were raised about data quality, transparency, and risks associated with AI.
- Participants identified a need for standardized tools and professional data management to support FAIR data.