greenwashing

AI-Augmented ESG Analysis

Attempting to create an LLM to detect corporate greenwashing in professional texts

Streamlit app that screens sustainability report PDFs for CSRD/ESRS readiness, ESG disclosure maturity, framework coverage (GRI, TCFD, GHG Protocol, SDG) and greenwashing risk. Rule-based, no paid APIs.

NLP-powered greenwashing scanner for fashion - paste a product link, get the True Eco-Score

An AI-powered Streamlit web app to detect greenwashing in sustainability claims using DistilBERT (MNLI) model. Classifies input text into: Greenwashing, Genuine, or Marketing Hype. Built by Team BharatWin 🇮🇳.

Lanterna verde é um sistema de avaliação do nível de práticas de greenwashing de uma organização ou produto.

code and documentation for my Bachelor's thesis, which delves into a data-driven exploration of sustainability reports from DAX 40 companies using Natural Language Processing (NLP). The analysis aims to uncover potential greenwashing tendencies through a comprehensive examination of sentiment values, SDG alignment, ESG scores, and company size.

NLP and machine learning system for detecting vague and unverifiable sustainability claims in corporate ESG reports, with rule-based scoring, TF-IDF classification, cross-sector analysis, and an interactive Streamlit dashboard.

NLP tool to flag greenwashing and provide insights through corporate reports

多智能体人工市场：政策与市场双轮驱动下的 ESG 转型与漂绿博弈 · Multi-agent artificial market for ESG disclosure, greenwashing and policy experiments (research in progress)

Analyzes 10 years of SEC 10-K filings from S&P 50 firms using NLP and Guided LDA to evaluate ESG communication clarity and detect greenwashing. Introduces the Greenwashing Risk Index (GRI) to measure balance between ESG emphasis and linguistic transparency.

Explainable greenwashing-risk scoring for corporate climate disclosures

Quantitative framework for detecting corporate greenwashing using ESG deficiency scores and portfolio optimization.

ML tool that cross-references company ESG claims against independent news and emissions data to detect greenwashing.

ClimateBERT-based greenwashing index for S&P 500 Energy & Utilities — MSc dissertation, TCD 2026

Verified, source-by-source research on the energy, emissions, water and grid impacts of AI and data centres, and the corporate claims made about them.

A framework-agnostic Jev decision layer for ESG reporting — reads every claim in a sustainability report against the evidence behind it. Works over GRI, ESRS, BRSR or ISSB. Zero dependencies, 22 tests.

Evidence-grounded ESG claim extraction and auditing: deterministic LLM post-correction with offline, reproducible evaluation

Data from thesis on how to detect greenwashing in a semi-automated way

NLP-based greenwashing risk screener for Indian BRSR ESG disclosures (FY2022-25). 15 real public filings, 5 sectors. Python, SQL, Excel, FinBERT.