AI Agents for Sustainable SMEs: A Green ESG Assessment Framework

A new research paper from authors Viet Trinh, Tan Nguyen, Minh-Huyen Phan, and Quan Luu introduces a scalable AI-driven framework for assessing Environmental, Social, and Governance (ESG) performance in European small and medium-sized enterprises (SMEs). The study, submitted to arXiv on April 5, 2026, operates in two phases. First, expert-validated ESG baseline scores were established from a subset of the Flash Eurobarometer FL549 survey data. Second, an AI agent system built on the open-source n8n automation platform applies these baselines to automatically classify SMEs' ESG performance and produce contextual recommendations using large language models (LLMs). The results demonstrate that the AI system's outputs are highly consistent with human-derived assessments, suggesting that automated ESG monitoring can be both accurate and scalable.

The framework is specifically designed to support the European Green Deal's goals by enabling more effective monitoring and intervention strategies for SMEs, which often lack resources for comprehensive ESG reporting. By leveraging n8n — a workflow automation tool popular for connecting APIs and services — the researchers created a modular, low-code agent pipeline that can be adapted by regulators, consultants, or SME networks. The use of LLMs allows the system to generate actionable, context-aware recommendations beyond simple classification. While the paper is academic, the approach points to a practical path for democratizing ESG compliance in Europe's massive SME sector, potentially reducing reporting costs and improving sustainability outcomes at scale.