Skip to main content
 

Data-Driven EnviroLab launches AI-powered platform to track climate data

January 30, 2025 Chat NDC homepage.

Today marks the official launch of ChatNDC, an innovative platform designed to transform the analysis and understanding of global climate policies. Developed collaboratively by the Data-Driven EnviroLab and Arboretica, and supported by ClimateWorks Foundation, ChatNDC leverages advanced artificial intelligence to empower nations and stakeholders in their efforts to meet and surpass their commitments under the Paris Agreement.

By incorporating more than 800 authoritative scientific and policy documents on climate change and actions governments and non-government stakeholders are taking, ChatNDC uses AI to simplify the analysis of climate policies. Users can interact with the platform through a chatbot interface, asking questions related to climate action across eight key themes. It offers NDC overviews, detailing emissions targets, timelines, and priorities, along with emissions profiles that break down greenhouse gas contributions by sector. Users can follow progress toward targets, explore adaptation and resilience strategies, and find relevant climate finance sources on mitigation and adaptation efforts. The platform highlights international cooperation and partnerships, captures contributions from non-state actors such as businesses and NGOs, and identifies opportunities for greater alignment with global climate goals.

Angel Hsu, director of the Data-Driven EnviroLab and associate professor at UNC-Chapel Hill, states, “By streamlining access to climate policy data, ChatNDC empowers policymakers to bridge the gap between ambition and action. By putting comprehensive climate data at policymakers’ fingertips, we’re enabling them to identify credible opportunities for enhanced ambition and make more informed decisions. This tool has the potential to transform how we understand and implement climate policies, ultimately driving more effective global climate action.”

Key Features of ChatNDC include:

  • Detailed policy analysis and summaries
    ChatNDC provides detailed, country-specific insights into NDCs, allowing users to efficiently interrogate specific emissions reduction targets, timelines, and sectoral priorities. Users can further explore how a country’s NDC has changed over time in these specific categories. 
  • ‘Anti-hallucination’ and detailed references
    ChatNDC’s ‘anti-hallucination’ feature aims to minimize inaccuracies and fabrications commonly referred to as “hallucinations” in generative AI outputs. This feature involves sophisticated algorithms that cross-verify generated responses against a database of trusted sources in real time. When ChatNDC generates content, it simultaneously cites specific references, providing users with a detailed trail of source materials. This transparency ensures that users receive not only accurate but also verifiable information. 
  • Document-level exploration and data catalogue
    Users can chat specifically with individual documents used to train ChatNDC. Users can query specific parts of the dataset, and ChatNDC can provide responses directly from the data, highlighting the relevant sections of the text in the source document. This feature essentially turns the training dataset into an interactive exploration tool, where users can see exactly how information is being processed and presented by the AI chatbot.

“With ChatNDC, we are transforming the landscape of climate policy research by prioritizing transparency and accuracy in real-time data aggregation,” said James Zhang, CEO of Arboretica, the start-up leading the development of ChatNDC. “Our goal is to establish ChatNDC as the definitive one-stop solution for all climate policy research, ensuring that every stakeholder has instant access to the most reliable and up-to-date information available.”

With the World Economic Forum’s 2025 Global Risks Report identifying misinformation and disinformation as the most critical growing short-term risk for the second year in a row, credible tools that help users decipher accurate information are critical. 

In response to this need, DDL is planning the new Center for Climate Leadership and AI-driven Integrity (CLAIM), made possible by a planning grant from the National Science Foundation. CLAIM aims to harness the expertise of computer scientists, social scientists, law and policy experts, and practitioners from various fields to tackle the challenge of AI potentially propagating false information within climate action policies.

ChatNDC exemplifies the type of AI tool designed to align with CLAIM’s objectives by providing accurate, credible information through a robust mechanism for verifying and disseminating fact-checked information on climate policy. It is now available and can be accessed from chatndc.org

About the Data-Driven EnviroLab (DDL):

The Data-Driven EnviroLab (DDL) is an interdisciplinary and international group of researchers, scientists, programmers, and visual designers. DDL uses innovative data analytics to distill signals from large-scale and unconventional datasets and develop policy solutions to contemporary environmental problems. We promote evidence-based approaches to problem-solving while boosting information disclosure and transparency among public institutions, private companies, civic organizations, and individual citizens.

DDL is based at the Institute for Environment at the University of North Carolina Chapel Hill. As an academic research lab led by Angel Hsu, associate professor at UNC-Chapel Hill, we have a particular mission to help train data-minded scholars and leaders in the field of environmental policy. 

DDL has recently published articles in peer-reviewed journals such as One Earth, Environmental Research: Infrastructure Sustainability, Climate Policy, Nature Communications, and Nature Urban Sustainability. The group also publishes yearly reports on Global Climate Action in collaboration with NewClimate Institute and researchers at Utrecht University, the German Institute of Development and Sustainability (IDOS), Radboud University, and CDP.  DDL Director Angel Hsu was a contributing author to the IPCC 6th Assessment Report, lead author of the 2018 UNEP Emissions Gap Report chapter on non-state and subnational actors, and was recently named a lead author for the IPCC Special Report on Cities and Climate Change.

About Arboretica:

Arboretica, headquartered in Rotterdam (Netherlands), is a pioneer in using machine learning and natural language processing to automate the work of environmentalists and policymakers. They constantly innovate cutting-edge, peer-reviewed data-driven technologies to make sustainability analyses more streamlined and scalable.

Arboretica has helped global policymakers, universities, NGOs, and corporations to automate manual working processes, discover hidden insights, and create tangible impact on climate change, nature, and biodiversity. The team has led the creation of many AI-driven environmental solutions, including ChatNetZero and GreenSearch.AI

Media Contact:

Nicholas Pell, ddl@unc.edu.