Automating ESG Reporting with Generative AI
The growing importance of Environmental, Social, and Governance (ESG) reporting has become a significant operational burden for organizations worldwide. Companies are establishing baselines to track their progress, supported by an expanding framework of reporting standards, some mandatory and some voluntary. However, ESG reporting has evolved into a substantial challenge, with 55% of sustainability leaders citing excessive administrative work in report preparation, while 70% indicate that reporting demands inhibit their ability to execute strategic initiatives.
The Challenge of ESG Reporting
The growing complexity of ESG reporting has become a significant operational burden for organizations worldwide, with 55% of sustainability leaders citing excessive administrative work in report preparation and 70% indicating that reporting demands inhibit their ability to execute strategic initiatives.
To address this challenge, Gardenia Technologies developed Report GenAI, an agentic framework using generative AI models on Amazon Bedrock to automate large chunks of the ESG reporting process. Report GenAI pre-fills reports by drawing on existing databases, document stores, and web searches, and then works collaboratively with ESG professionals to review and fine-tune responses.
The Report GenAI Workflow
The Report GenAI workflow consists of five steps: setup, batch-fill, review, edit, and repeat. During setup, the agent is configured and authorized to access relevant data sources. The batch-fill step involves the agent iterating through each question and data point to be disclosed, retrieving relevant data from client document stores and document searches, and processing this information to produce responses in the expected format.
Key Components of Report GenAI
- Lightweight UI: Hosted on auto-scaled Amazon ECS Fargate, providing an interactive interface for users.
- Central Agent Executor: Uses the reasoning capabilities of leading text-based foundation models to break down user requests and orchestrate tasks.
- Web Search Tool: Retrieves content from public web pages to formulate responses.
- Text-to-SQL Tool: Generates and executes SQL queries to retrieve data from the company’s emissions database.
- Retrieval Augmented Generation (RAG) Tool: Accesses information from corporate document stores and uses it as a knowledge base.
Evaluating Agent Performance
To ensure accuracy and reliability in ESG reporting, Report GenAI implements a sophisticated dual-layer evaluation framework that combines human expertise with advanced AI validation capabilities.
Human Expert Validation
The solution’s human-in-the-loop approach allows ESG experts to review and validate AI-generated responses, providing transparency into the agent’s decision-making process.
AI-Powered Quality Assessment
Report GenAI uses state-of-the-art LLMs on Amazon Bedrock as LLM judges to evaluate response accuracy, completeness, and consistency with provided context.
Results: 75% Reduction in Reporting Time
Omni Helicopters International cut their CDP reporting time by 75% using Gardenia’s Report GenAI solution, reducing their reporting time from one month to just one week.
In conclusion, Report GenAI is available on the AWS Marketplace today, offering a powerful solution to automate ESG reporting and free up resources for more strategic sustainability initiatives.
