Third place AIM Hackathon: Dissecting ESG reports by relevance with LLMs

Authors: Andrei I. Cursaru and Orion Forowycz

The Venturers team created an architecture to identify the relevance of each section. Their prototype classified sentences based on its content, making it easier to identify concrete facts versus promotional, vague statements. This relevance map helps users quickly navigate reports, enabling standardized comparisons between companies and incentivizing clarity in ESG reporting.

Dissecting ESG reports by relevance with LLMs

ESG (Environment, Social, Governance) reports made yearly by companies are initially thought of as a way for investors to be able to easily assess how investing in a given company may have indirect environmental or social negative impacts.

However, the interests of the people writing these reports are not aligned with the interests of the people reading them. For companies, the goal is to make themselves look good and to make it as tedious as possible for readers to easily and quickly find the hard facts they are looking for.

As a result, ESG reports drown the facts in an ocean of vague and often meaningless motivational statements about the company’s beliefs, values, long-term goals, and so on. For investors, these digressions carry little to no information about the potential negative impacts of a company, since no company will ever claim that they strive to destroy the planet and mistreat all their workers.

How could we use AI to help us to get more use out of these reports?

Large language models (LLMs) like GPT-4o can do many things, but they can’t directly tell us if the information that a company reports is true, partial, biased or irrelevant. They can’t simply tell us if and how much greenwashing fills the pages of an ESG report, because the information to precisely do so is not out there.

During the hackathon Sustainability meets LLMs organised by AIM – AI Impact Mission and supported by TIMETOACT Group Österreich, the team The Venturers came up with a simple idea to make ESG reports more useful: They process each sentence of a report with an LLM to evaluate its relevance and usefulness to investors, in order to generate a "relevance map" of the report and summary statistics of relevance, and to be able to compare reports of different companies.

The prototype parses a PDF document page by page and sentence by sentence (including figures provided) and feeds it to a model (in our tests, GPT-4o) via the respective API. The model is prompted to answer a set of pre-defined questions by yes or no, such as:

  • "Does this statement or figure say something about the values or beliefs of the company?"
  • "Does this statement or figure state a quantified, concrete fact about the company?"

This output is then used to classify the string in one of more categories: beliefs & values, goals and missions, quantified hard facts about the company, qualitative facts about the company, facts unrelated to the company.

By combining these outputs with the length of the statement, they were able to easily infer percentages of facts vs abstract statements for each page, report, or company. This can then be used to:

  • Create a "relevance map" of the PDF, which outlines for example that facts about the company's environmental progress are only on pages 35-39 out of a 100+ page PDF. This can guide investors to quickly read only the relevant parts of the report, without relying on the report's often not so informative table of contents.

  • Extract only statements of a desired category into a new document, for example quantified hard facts about the company.

  • Create conciseness scores for reports based on these proportions.

  • Compare different companies to each other in a standardised way in terms of how much they try to drown the facts in their reports with decorative text. On the long term this could help to hold companies accountable for clear and readable reports, by giving them an incentive to increase their conciseness score and focus their reports on facts.

 

Outlook:

The prototype could be improved in various ways, for example by fine-tuning the categories and making more of them, depending on the sections of the report of the type of facts which are stated. Using GPT-4o is convenient for prototyping and for a good understanding of the context, but is likely rather overkill on the longer for this task. Working on using a smaller but still efficient model would improve the solution by reducing costs, energy consumption and the carbon footprint overall.

Blog 11/5/24

AIM Hackathon 2024: Sustainability Meets LLMs

Focusing on impactful AI applications, participants addressed key issues like greenwashing detection, ESG report relevance mapping, and compliance with the European Green Deal.

Blog 10/30/24

Second Place - AIM Hackathon 2024: Trustpilot for ESG

The NightWalkers designed a scalable tool that assigns trustworthiness scores based on various types of greenwashing indicators, including unsupported claims and inaccurate data.

Blog 11/4/24

SAM Wins First Prize at AIM Hackathon

The winning team of the AIM Hackathon, nexus. Group AI, developed SAM, an AI-powered ESG reporting platform designed to help companies streamline their sustainability compliance.

Unternehmen 1/19/23

Sustainability in the TIMETOACT GROUP

Sustainability is one of the big topics of our time and we also want to get involved and face up to our responsibility as TIMETOACT GROUP. Find out everything about our sustainability activities here.

Blog

How I Won the Enterprise RAG Challenge

In this article, Ilia Ris describes the approach that helped him achieve first place in both prize categories and the overall SotA leaderboard.

Headerbild zur Third Party Integration
Technologie

Third Party Integration

The concept of bringing together the most important information in a central Digital Workplace platform (HCL and beyond) is now being continued with the ICEC Atlassian Confluence & Jira integration.

Wissen 8/30/24

LLM-Benchmarks August 2024

Instead of our general LLM benchmarks, we present the first benchmark of different AI architectures in August.

Blog 8/11/22

Part 1: TIMETOACT Logistics Hackathon - Behind the Scenes

A look behind the scenes of our Hackathon on Sustainable Logistic Simulation in May 2022. This was a hybrid event, running on-site in Vienna and remotely. Participants from 12 countries developed smart agents to control cargo delivery truck fleets in a simulated Europe.

Kompetenz

Governance & Operational Excellence

Digitalization inevitably leads to business processes changing and roles and responsibilities being redistributed. At the same time, new legal requirements, regulations and standards need to be met.

Insights

Team-Leaderboard of the Enterprise RAG Challenge

The team-leaderboard includes all submitted entries – including those submitted after the Ground Truth was released. Therefore, we consider this ranking an unofficial overview.

Sustainability. Wir unterstützen Sie dabei, Ihre Nachhaltigkeitsstrategie zu entwickeln und umzusetzen.
Service

Sustainability

We support you in developing and implementing your sustainability strategy.

Wissen 4/30/24

LLM-Benchmarks April 2024

This LLM Leaderboard from april 2024 helps to find the best Large Language Model for digital product development.

Neues Training für Atlassian Jira: „Realizing the Power of Jira Reporting & Dashboards” jetzt auch für die Cloud
Training

Realizing the Power of Jira Reporting & Dashboards (DC)

Over the course of "Realizing the Power of Jira Reporting & Dashboards (Data Center)" participants will learn about the basic functions of Jira dashboards and reports.

Wissen 7/30/24

LLM-Benchmarks July 2024

This LLM Leaderboard from July 2024 helps to find the best Large Language Model for digital product development.

Blog

Your secure solution for the digital working world

Google Workspace: The secure alternative in today's threat landscape!

Insights

These are the proud winners of the Enterprise RAG Challenge

Discover the winners of the Enterprise RAG Challenge! Explore top RAG solutions, watch the official announcement, and see how AI-driven retrieval and LLMs shaped the best-performing models.

Headerbild zu Mendix in der Fertigungsindustrie
Technologie

Mendix in the manufacturing industry

Have solutions designed directly by the experts in your company's divisions and thus ensure that the solution implements the exact requirements of the specialist department. Use a wide variety of connectors to collect and evaluate all decision-relevant information.

Wissen 6/30/24

LLM-Benchmarks June 2024

This LLM Leaderboard from june 2024 helps to find the best Large Language Model for digital product development.

eXplain – Train The Boss Whitepaper
Whitepaper 7/2/24

eXplain – Train the Boss Whitepaper Download

eXplain – Finally, an end to the legacy dilemma! No more „That‘s not possible, we‘ll do it later, it‘s far too time-consuming“

Junger Business Mann der seinen Erfolg feiert
Event

We are 2024 Google Cloud Sales Partner of the Year - Alps!

We are proud to announce that we have won the prestigious 2024 Google Cloud Sales Partner of the Year - Alps Award!

Bleiben Sie mit dem TIMETOACT GROUP Newsletter auf dem Laufenden!