IFC assesses ESG risk with AI-powered MALENA platform
The International Finance Corporation (IFC), part of the World Bank Group, built MALENA (Machine Learning ESG Analyst), an AI-powered platform that uses NLP, sentiment analysis and named entity recognition to extract ESG insights from unstructured text such as impact assessments, news articles and sustainability reports at scale. Built on the Databricks Data + AI Platform with GPU computing and MLflow for the ML lifecycle, MALENA analyzes 19,000 sentences per minute (a 950x increase over manual review) and identifies 1,200 ESG risk terms, reducing document analysis time from weeks to days and enabling IFC to offer MALENA to external users as the World Bank Group's first AI as a Service.
Overview
The International Finance Corporation (IFC), part of the World Bank Group, built MALENA (Machine Learning ESG Analyst), an AI-powered platform that uses NLP, sentiment analysis and named entity recognition to extract ESG insights from unstructured text such as impact assessments, news articles and sustainability reports at scale. Built on the Databricks Data + AI Platform with GPU computing and MLflow for the ML lifecycle, MALENA analyzes 19,000 sentences per minute (a 950x increase over manual review) and identifies 1,200 ESG risk terms, reducing document analysis time from weeks to days and enabling IFC to offer MALENA to external users as the World Bank Group's first AI as a Service.
This entry has 13 published fields tied to exact passages in an immutable source capture.
Inspect the highlighted sourceThe challenge
Previously reliant on manual processes and limited data processing tools, the operationalization of ML models to analyze roughly 21 million pages of text took months to complete, and training and fine-tuning large language models was limited and resulted in frequent failures, with limited cluster choices impacting the ability to optimize model performance.
The solution
IFC built MALENA, an AI-powered platform that supports the review and analysis of massive amounts of text to identify ESG opportunities and risks; IFC has trained sentiment analysis models to extract insights from across hundreds of thousands of documents, and named entity recognition is used to integrate external data sources and connect news stories to client profiles, all running on the Databricks Data + AI Platform with GPU computing and MLflow for the ML lifecycle.
Reported business value
In-house document analysis times have been reduced from weeks to days for ESG due diligence; MALENA can analyze 19,000 sentences per minute, compared with the average human reader who can only read 15 to 20 sentences per minute, a 950x increase; and IFC has developed 10,000 company profiles and expanded insights for more than 180 markets.
Sources
Open any source and check the claim yourself — that is the point of the register.
This record was researched and written with AI assistance, and its claims were checked against the sources above. (EU AI Act art. 50 transparency notice.)
Other government & public sector entries in the register.
GovTech unlocks data insights to improve nationwide services with Databricks
Singapore's Government Technology Agency (GovTech) migrated from an on-premises dashboarding system to the Databricks Data + AI Platform on AWS with Unity Catalog and Delta Lake, cutting dashboard creation time from 90 to 30 days, democratizing data across 50% of corporate divisions in the first year, and saving 8,000 labor hours annually.
GovTech· SingaporeDelta Lake · Lakeflow Connect · Lakeflow Jobs +1Estonia rolls out Bürokratt, an AI-guided virtual assistant network for public services
Bürokratt is a network of chatbots deployed on Estonian public sector institutions' websites, letting people obtain information from institutions and use public and information services via virtual assistants. It is a state-created, AI-based digital assistant that helps institutions deliver modern, efficient, around-the-clock customer service using large language models.
VA Advances Healthcare Insights With AI
The U.S. Department of Veterans Affairs leverages Databricks Data Intelligence to modernize healthcare analytics for millions of veterans, unifying massive distributed datasets into a single secure environment and streaming petabytes of health data in real time, reducing processes that once took hours to seconds. Databricks enables AI and large language models to detect risk early and enhance governance.
U.S. Department of Veterans AffairsDatabricksAustrian Academy of Sciences unlocks Ancient Greek with Mistral
The Austrian Academy of Sciences (OeAW), together with its Austrian Archaeological Institute, partnered with Mistral and services partner Reply to build Apollo, described as the first advanced large language model for Ancient Greek. Apollo is trained on a specialized corpus of 600 million words of historical Greek text plus tens of thousands of published inscriptions and papyri, helping researchers reconstruct damaged texts and identify thematic connections across collections. The OeAW reports Apollo turns work that once took years into hours, addressing over one million unread Greek papyri worldwide, with future phases planned for semantic search and handwritten inscription decipherment.
Austrian Academy of Sciences· AustriaMistralWas this helpful?
Your feedback helps us improve our use case database