EcoLens GRIE: AI-Driven Detection of Greenwashing Claims Using Transformer-Based NLP

Abstract

The rapid growth of e-commerce and digital marketing has significantly increased the number of environmental claims made by companies regarding the sustainability of their products. While many organizations promote environmentally friendly initiatives, a large number of these claims lack verifiable evidence and are often misleading. This practice, commonly known as greenwashing, creates confusion among consumers and undermines trust in genuine sustainability efforts. Identifying deceptive environmental claims has therefore become an important challenge in modern digital commerce. In this research, we present EcoLens GRIE (Greenwashing Risk Intelligence Engine), an Artificial Intelligence-based system designed to detect misleading environmental claims using Natural Language Processing and machine learning techniques. The proposed system analyzes sustainability claims from product descriptions, marketing materials, and online listings to evaluate their credibility and transparency. A transformer-based zero-shot classification model is used to identify patterns associated with common greenwashing practices such as vague language, hidden trade-offs, and absence of supporting evidence. The system computes multiple analytical metrics including deception probability, transparency index, credibility score, and environmental confidence. These metrics are combined to generate a comprehensive sustainability intelligence report for consumers. The EcoLens platform is implemented using a FastAPI backend integrated with transformer-based NLP models and a React-based interactive dashboard for visualization of results. Experimental analysis demonstrates that the proposed system can effectively identify misleading sustainability claims and provide real-time risk assessment. The system not only assists consumers in making informed purchasing decisions but also encourages companies to adopt transparent environmental communication practices.

Introduction

The text addresses the growing issue of greenwashing, where companies mislead consumers by making vague or unverified environmental claims. As sustainability becomes a major concern, many consumers prefer eco-friendly products, but the rise of digital marketplaces makes it difficult to verify such claims. Traditional detection methods (like manual audits) are slow and cannot keep up with the volume of online content.

To solve this, the study proposes EcoLens GRIE, an AI-driven system that uses Natural Language Processing (NLP) and machine learning to automatically detect misleading sustainability claims. It analyzes marketing text using transformer-based models, zero-shot classification, and rule-based evaluation (such as the “Seven Sins of Greenwashing”) to assess credibility.

The system works through a multi-stage pipeline:

• Data collection using advanced web scraping techniques

• Neural analysis to classify industries and detect deceptive patterns

• Scoring metrics including credibility score, deception probability, and transparency index

• Feature extraction and clustering to identify patterns and trends in greenwashing across industries

EcoLens GRIE can process large volumes of data in real time, making it more scalable than traditional methods. It also helps consumers make informed decisions while encouraging companies to be more transparent.

Results show that the system effectively distinguishes between genuine and misleading claims—vague statements tend to score high in deception, while specific, evidence-based claims score higher in credibility.

Conclusion

The EcoLens Greenwashing Risk Intelligence Engine (GRIE) project has successfully demonstrated that neural-based auditing is a powerful and necessary tool for modern environmental protection. By integrating advanced natural language processing, zero-shot classification, and a resilient scraping infrastructure, the platform provides a transparent and objective method for identifying deceptive sustainability claims. Our research shows that semantic similarity-based machine learning can effectively map corporate declarations against established frameworks like the \"Seven Sins of Greenwashing,\" offering consumers a reliable metric for credibility in an increasingly cluttered digital marketplace. The modular architecture ensures the system is ready for future integration with decentralized supply-chain data and diverse linguistic markers. The findings of this study emphasize that the detection of greenwashing must move beyond superficial keyword analysis into the realm of deep semantic understanding. The success of the `BART-large-MNLI` model in identifying risk markers without extensive domain-specific training highlights the scalability of zero-shot approaches in fast-evolving fields like sustainability. Furthermore, the development of an Eco-Resilient Scraper showcases the importance of addressing technical barriers to information access, ensuring that auditing tools remain effective even against aggressive corporate anti-bot measures. This methodology provides a blueprint for future AI-driven consumer protection platforms across multiple domains. In conclusion, EcoLens GRIE empowers consumers to make more environmentally conscious choices while holding corporations accountable for their marketing claims. By providing actionable indices such as the Credibility Score and Deception Probability, we are fostering a more transparent and honest green marketplace. Future work will focus on expanding the system to support multi-modal analysis, including image-based greenwashing detection and integration with blockchain-based verification systems. As the global community strives towards genuine sustainability, tools like EcoLens GRIE will be essential for ensuring that environmental Progress is measured by actual performance rather than perceived marketing.

References

[1] S. V. de Freitas Netto, M. F. Sobral, A. R. Ribeiro and G. R. da Luz Soares, “Concepts and forms of greenwashing: A systematic review,” Environmental Sciences Europe, vol. 32, no. 1, pp. 1–12, 2020. [2] M. A. Delmas and V. C. Burbano, “The drivers of greenwashing,” California Management Review, vol. 54, no. 1, pp. 64–87, 2011. [3] TerraChoice Environmental Marketing, The Seven Sins of Greenwashing, TerraChoice Group Inc., Canada, 2010. [4] S. Szabo and J. Webster, “Perceived greenwashing: The effects of green marketing on environmental and product perceptions,” Journal of Business Ethics, vol. 171, pp. 719–739, 2021. [5] J. Devlin, M. W. Chang, K. Lee and K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding,” in Proc. NAACL-HLT, Minneapolis, USA, 2019, pp. 4171–4186. [6] A. Vaswani et al., “Attention is all you need,” in Proc. Advances in Neural Information Processing Systems (NeurIPS), 2017, pp. 5998–6008. [7] N. Reimers and I. Gurevych, “Sentence-BERT: Sentence embeddings using Siamese BERT networks,” in Proc. EMNLP, 2019, pp. 3982–3992. [8] Y. Yin, X. Zhou, W. He, L. Liu and C. Li, “Benchmarking zero-shot text classification: Datasets, evaluation and entailment approach,” in Proc. EMNLP, 2019. [9] S.S.D.K. Maha Lakshmi, Umamaheswararao Mogili, Sravya Eluri, Dogga Ramachandra Rao. (2023), “Online Dynamic Out Patient Queue System for Automated Token Generation in Hospitals”, Science, Technology and Development Journal, Volume XII, Issue VII, pp 71-78, DOI:23.18001.STD.2023.V12I07.23.37707. [10] M. Lewis et al., “BART: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension,” in Proc. ACL, 2020. [11] K. Ong, R. Mao, D. Varshney and E. Cambria, “Robust ESG analysis against greenwashing risks using artificial intelligence,” arXiv preprint arXiv:2502.15821, 2025. [12] Mogili, U., & Mohamed, A. (2023, November). Artificial intelligence and machine learning in the fields of education, medical, and smart phones. In AIP conference proceedings (Vol. 2917, No. 1, p. 050012). AIP Publishing LLC. [13] Mogili, U., Ampolu, K. V., Rajasekharam, B., & Timothy, M. J. AI-Driven Interaction in AR Environments, in Journal of Digital Economy, 2024, Volume 3, Issue 1, pp. 228-234. [14] Timothy, M. J., Rajasekharam, B., Ampolu, K. V., & Mogili, U. Threat Detection Using AI in Cybersecurity Systems, in IJIS, 2023, Volume 7, Issue 1, pp. 1-7. [15] Ampolu, K.V., Mogili, U., Timothy, M. J., & Rajasekharam, B. Machine Learning Models for Predictive Maintenance, in IJIS, 2022, Volume 6, Issue 4, pp. 1-7. [16] Rajasekharam, B., Timothy, M. J., Mogili, U., Ampolu, K.V., Machine Learning Models for Predictive Maintenance, in JDE, 2023, Volume 2, Issue 2, pp. 95-101. [17] Soujania, B., Ampolu, K. V., Timothy, M. J., & Mogili, U. (2025) Classifying Disease Information Forums through Semantic Similarity-Based Machine Learning, Science, Technology and Development Journal, Volume XIV, Issue II, pp 67-75. [18] B Satish Kumar, Kavitha C., Mogili, U.R., S. Pallam Shetty (2022). “Application of Machine Learning To Enhance the Performance of The Prophet Routing Protocol For Delay Tolerant Networks”. Journal for Basic Sciences, Volume 23, Issue 5, 2107-2116, DOI:10.37896/JBSV23.5/2278. [19] Sree Geeta, Umamaheswararao Mogili. (2022), “Use of Several Machine Learning Algorithms for Effective Prediction of Cyberbullying”, International Journal of Creative Research Thoughts, Volume 10, Issue 6, pp 17.

Copyright

Copyright © 2026 Umamaheswararao Mogili, M. Sahitya, V. Leelakrishna, K. Chandini, K. Mohini, K. Venkataramana. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.