2026 July ESIP Meeting

AI capabilities are developing tremendously quickly, creating opportunities for those supporting the research data enterprise to do their work in new or more efficient ways. But the rapid pace also creates divides: some people have access to the time, expertise, and resources to learn how to apply AI tools, while others do not. This session (Part 2 of the AI curation series) shares real-world experiences and lessons learned about how people are applying specific AI tools to the work of managing and making research data useful. Lightning talks followed by a roundtable discussion covering what is working well, what has been tried but is not ready for prime time, and key considerations and cautions.

Presentations
Dave Blodgett: Experiences with agent-based AI for software package and reproducible data pipeline development
Mike O’Donnell: Integrating Scientific Integrity into AI‑Driven Literature Synthesis
Carolina Berys: From Cruise Reports to Structured Metadata: What We Learned Using LLMs at CCHDO
Ryan Berkheimer: NOAA Foundations for Cognitive AI
Althea Marks: A FAIR Data Pipeline: From Pipe Dream to Reality with GenAI
Jordan Read: From Financial Ledgers to Dataset Citations: CUAHSI's Use of AI Across a Data Management Spectrum

Audience
Those with a success story or cautionary tale about applying AI to research data management; those who want to understand how AI might help address their own challenges and tasks