AI agents are now capable of searching for evidence, using tools and completing multi-step scientific tasks. But for those agents to work effectively in agriculture, the research they rely on has to be structured in ways automated systems can actually use — and much of it currently isn’t.

In an upcoming webinar hosted by the Council for Agricultural Science and Technology (CAST), researchers from Cornell University’s College of Agriculture and Life Sciences will introduce FAIR4AG², a framework that extends the well-known FAIR data principles for the era of AI agents. Luke Qian, postdoctoral associate in food science, and Martin Wiedmann, the Gellert Family Professor in Food Safety, will define what “agent-actionability” means in practice — whether an agent can find, access, execute, trace and interpret a research resource without human intervention — and offer a roadmap for making agricultural science more accessible to trusted AI systems.

Topics include structured document formats, machine-readable metadata and licensing, APIs, Model Context Protocol (MCP) servers, Agent Skills and reproducible runtime records. The session will also address how to maintain easy access and understanding of research materials by human readers alongside improvements aimed at AI systems.

The webinar is free and open to all.

Sept. 22, 2026 | 12:00–1:00 p.m. CT | Free