Agents for Experiments, Experiments for Agents: A Design Grammar for AI-Enabled Experimental Science

TL;DR

This paper introduces SEED, a framework using actor-flow graphs to represent, evaluate, and generate experimental conditions for AI and multi-agent systems, enhancing transparency and governance.

Contribution

It presents SEED, a novel design grammar for experimental workflows that improves traceability, comparison, and governance of AI-enabled experiments.

Findings

SEED-guided designs show clearer actor-flow changes and governance checks.

The framework supports evaluating structural novelty of experimental conditions.

Feasibility tests indicate SEED's effectiveness as a design aid.

Abstract

AI systems are becoming active participants in organizational and knowledge work. They increasingly interact with humans, coordinate workflows, and operate in multi-agent arrangements. Understanding their effects therefore requires more than measuring output accuracy; it requires evidence about mechanisms, delegation, feedback, and control. Experiments remain central to this task, but they also face a recursive challenge: we need experiments for agents to study these arrangements, and we may need agents for experiments to help search the expanding space of possible designs. Yet experimental conditions for human-AI and agentic workflows are still largely specified in prose, making them difficult to compare, reuse, or audit. We frame this as a problem of workflow representation, traceability, and governance in AI-enabled knowledge production. We introduce SEED (Structural Encoding for Experimental Discovery), a framework that represents experimental conditions as typed actor-flow graphs. SEED supports three design functions: describing conditions as interaction structures, evaluating structural novelty relative to encoded prior designs, and generating candidate designs under feasibility and governance constraints. We report a lightweight empirical feasibility test that compares graph-blind and SEEDguided generation in a medical-triage design task. In this diagnostic contrast, SEED-guided candidate designs show clearer actor-flow changes, assumptions, and governance checks, supporting the feasibility of the grammar as a design aid. The commentary closes by identifying governance tensions around novelty, replication, validity, diversity of inquiry, and accountability.