U.S. team: Differentiating greenwashing from sincere corporate efforts towards ESG goals
Updated: May 2
U.S. Young Professionals, April 2025
Problem: How can we differentiate “greenwashing” from sincere corporate efforts to promote environmental, social, and governance (ESG) goals?
Solution: Use Natural Language Processing (NLP) to verify company’ claims
The U.S. Young Professionals Team explored how artificial intelligence can evaluate corporate reports of environmental, social, and governance (ESG) efforts. As government regulations increase the volume of corporate reporting, that raises the risk that companies may be providing misleading or inaccurate information, or “greenwashing.”
The team argued that the future of transparency lies in using Natural Language Processing (NLP) to verify company claims. They proposed two specific steps to strengthen AI's accuracy: first, train models on massive datasets of corporate jargon, and second, deploy targeted searches to find passages relevant to specific ESG risks. While basic NLP can struggle with context, the team recommended using advanced models like BERT and RoBERTa. These models can be fine-tuned on specific ESG pillars, thereby allowing the AI to detect subtle framing patterns and flag vague claims that lack accountability.
By encouraging greater consistency and standardization in reporting, these AI tools can effectively detect greenwashing language and increase stakeholder engagement. Furthermore, this technology offers greater accessibility for smaller companies to make their reports transparent. For GEN, harnessing these models strengthens our ability to evaluate small to medium-sized firms and identify partners that truly reflect our mission and core values.
Read more below:
Comments