What does it mean for scientific research to be truly reusable — not just by humans, but by AI agents? In this webinar, Drs. Martin Wiedmann and Luke Qian from Cornell University’s College of Agriculture and Life Sciences introduce FAIR4AG², a framework that extends the FAIR data principles to make research knowledge actionable for AI systems.
Wiedmann and Qian walk through why agricultural science lags behind biomedical research in AI accessibility, what the 77% reproducibility failure rate in biology tells us about how we publish, and why AI agents struggle with the same models that already exist in GitHub repositories and PDF papers. Their answer: knowledge units — packages of executable models, data, and interpretive instructions — delivered as plugins that connect directly to ChatGPT or Claude.
The session closes with a live case study: a milk spoilage decision-support plugin that allows a dairy processor to ask questions about spoilage causes, intervention strategies, and return on investment — with answers grounded in published models rather than generic AI responses.
