Empathy in the Soil, Agents on the Desktop: Why Plant Pathology Needs a Scientist-in-the-Loop AI Paradigm

Checked 25 Oct 2026. Non-commercial or no-derivatives license: full text kept internally; only metadata and abstract are indexed and used.

Plant pathology faces a cognitive bottleneck caused by the growing volume of biological data. Addressing it requires a shift from passive chatbots to autonomous AI research agents that operate in Goal-Action-State cycles. Researchers can set up a local, desktop-based agentic framework that combines semantic discovery, LLM-Wiki knowledge synthesis, and in silico prevalidation, managed under a human-directed three-tier organizational model. Strict "Scientist-in-the-Loop" boundaries, which means that a human researcher sets the goals, reviews AI-generated output before it is treated as evidence, and approves any consequential action, preserve human academic rigor and control, so plant pathologists can leave routine data processing to the system and focus on creative hypothesis design and physical validation.

• Three independent 2026 efforts converge on 'scientist-in-the-loop' as the pragmatic deployment form for research agents: Jeon 2026 prescribes strict boundaries in plant pathology under which a human sets goals, reviews AI output before it is treated as evidence, and approves consequential actions; Zhao et al. 2026 advocate 'pragmatic scientist-in-the-loop co-pilots' for drug discovery; and Huynh et al. 2026 built the drug-repurposing system RepurAgent around a human-in-the-loop design. Trace →
• The infrastructure barrier behind the democratization tension is being routed around at the small end: where Qi et al. 2026 warn that long-context biological workflows exceed typical academic or clinical infrastructure, Jeon 2026 has researchers set up a local, desktop-based agentic framework for plant pathology combining semantic discovery, LLM-Wiki knowledge synthesis and in silico prevalidation. Trace →

• AI for science and agentic research