Climate and Political Fact-Checking Research & Development Collaboration
CLIMATE+TECH is an AI Think-Tank bridging the gap between research and local stakeholders in climate change adaptation and democratic discourse, democratizing knowledge using AI technology. It stays up-to-date with research and has a focus on Climate and Political Claims.
Our Mission
CLIMATE+TECH is democratizing knowledge through AI technology, creating tools that:
- Verify Climate Claims: Check statements about climate change against trustworthy sources like IPCC reports
- Verify Political Claims: Fact-check political statements and policy claims against authoritative sources and evidence
- Detect Greenwashing: Identify and verify green claims and potential misleading greenwashing
- Combat Disinformation: Provide journalistic fact checks for both climate and political disinformation
Core Research Team
Interdisciplinary expertise from academia and industry
UZH Professor, Academic Advisory Lead - Academic Research
BI Norwegian Business School, Max Planck Institute for Human Development
C+T Managing Partner, Architecture
C+T Researcher, AI Engineers - AI Engineering
C+T Advisor, AI Expert - AI Advisory
XplAINLP Group Lead TU Berlin
C+T Researcher
Academic Advisor - Domain Expert
Technische Universität Berlin - ClimateCheck Research
SIRIS Lab, Universitat Pompeu Fabra - ClimateCheck Research
Data for good - ClimateCheck Research
DFKI - ClimateCheck Research
BIFOLD - ClimateCheck Research
Research Team & Collaborators
Our interdisciplinary team brings together expertise from academia and industry.
The Open Source Toolkit
Our Open Source Multi-Purpose toolkit serves researchers, developers, and the broader community:
🔬 For Researchers
- Benchmarking platform for new fact-checking algorithms
- Standardized datasets for evaluation and comparison
- Collaborative research environment with academic institutions
🛠️ For Developers
- Development tools for building fact-checking applications
- Configurable architecture for different use cases
- Integration capabilities with existing systems
🌍 For the Community
- Open source approach ensuring transparency and accessibility
- Community-driven development with stakeholder input
- Real-world applications serving diverse needs
Use Cases & Applications
| Use Case | Description | Target Users | Key Features |
|---|---|---|---|
| Climate Claims Verification | Automated verification of climate-related statements against authoritative sources like IPCC reports, ensuring accuracy in climate communications. | Researchers, Scientists, Policy Makers | IPCC integration, Source validation, Accuracy scoring |
| Greenwashing Detection | Advanced algorithms to identify and analyze potentially misleading environmental claims, helping combat corporate greenwashing. | Investors, Consumers, NGOs | Corporate claim analysis, Misleading pattern detection, Transparency scoring |
| Political Claims Verification | Automated verification of political statements, policy claims, and electoral information against authoritative sources and fact-checking databases. | Journalists, Media Organizations, Citizens, Policy Analysts | Real-time fact checking, Political statement analysis, Policy verification, Bias detection |
| Political Disinformation | Providing journalistic fact checks for political disinformation and misinformation campaigns. | Journalists, Media Organizations, Citizens | Real-time fact checking, Disinformation pattern detection, Source credibility analysis |
| Media Analysis | Tools for journalists and media organizations to fact-check climate-related content efficiently and accurately. | News Organizations, Content Creators, Publishers | Bulk content analysis, Editorial workflows, Citation verification |
Technical Approach
Our toolkit implements the Advocate Mediator Pattern against climate and political fact-checking datasets, using what we call the “Climinator pattern” for robust fact verification across multiple domains. The architecture is:
- Configurable and modular for different applications
- Scalable for various deployment scenarios
- Explainable to ensure transparency in fact-checking decisions
- Community-extensible for continuous improvement
Research Collaboration
We collaborate with leading academic institutions including:
- University of Zurich (UZH) - Led by Professor Markus Leippold
- ETH Zurich - Technical research partnerships
- Technische Universität Berlin (TU Berlin) - ClimateCheck research collaboration
- German Research Center for Artificial Intelligence (DFKI) - ClimateCheck systems development
- Universitat Pompeu Fabra - ClimateCheck research collaboration
- Industry partners - Real-world implementation and testing
ClimateCheck Collaboration
We are actively collaborating on the ClimateCheck initiative, a comprehensive research effort focused on scientific fact-checking of climate claims:
- ClimateCheck Dataset: Human-annotated dataset connecting 435 unique climate-related social media claims to 3,048 scientific abstracts
- Shared Tasks: Participation in ClimateCheck@SDP 2025 and ClimateCheck 2026 competitions
- Efficient Systems: Research on fine-tuned BERT models that achieve competitive results while being significantly faster than large language models
- Open Source: Contributing to open-source tools and datasets for the climate fact-checking community
Key Research Contributions:
- Development of efficient claim verification systems (0.032s per claim vs 0.363-12.229s for LLMs)
- Hybrid retrieval pipelines combining sparse and dense methods
- Fine-tuned models achieving competitive F1 scores (0.683) while being 11-380x faster
Resources:
- ClimateCheck Dataset Paper
- LLM vs BERT Comparison Research
- ClimateCheck 2025 Competition
- ClimateCheck 2026 Competition
Next Steps & Development Roadmap
Our ongoing development focuses on:
External Knowledge Integration
Like Wikipedia knowledge bases and other authoritative sources to expand our verification capabilities.
Large Fine-tuned Dataset Development
Building comprehensive benchmarking datasets for standardized evaluation of fact-checking systems. This includes collaboration on the ClimateCheck dataset, which connects social media claims to scientific literature.
Knowledge Graph Integration
Connecting current research and developing sophisticated knowledge representation systems.
Real World Applications
Deploying expert fact checkers in production environments for practical impact.
Community & Funding
Securing funding for core contributors and building a sustainable open source community.
Community Contributions
We’re looking for diverse contributions:
- Use cases and feedback from practitioners
- Datasets for training and evaluation
- Technical contributions to the open source codebase
- Research collaboration with academic institutions
- Stakeholder engagement for real-world validation
Get Involved
Whether you’re a researcher, developer, journalist, climate advocate, or political analyst, there’s a place for you in our community. Help us build the future of climate and political fact-checking!
Contact Information:
- GitHub: climateandtech/factchecker
- Slack: climateandtech.slack.com
- Email: contact@climateandtech.com
Ready to contribute? Contact us and let us know how you’d like to help democratize climate and political knowledge through AI.
02/2024: Literature Review
Partnerships and first configurable architecture
03/2024: Partnership with UZH / ETH
Collaboration with Markus Leippold and other researchers
06/2024: Implementation of Advocate Mediator Pattern
Against climate feedback dataset (Climinator pattern)
11/2024: Launch and Growing Community
Community readiness with use cases for media analysis and community notes
02/2025: Open Source - Architecture Improvement
More data sources, datasets, and web frontend development
Q2 2025: External Knowledge Integration
Wikipedia knowledge bases and authoritative sources integration
Q3 2025: Large Fine-tuned Dataset Development
Comprehensive benchmarking datasets for standardized evaluation
07/2025: ClimateCheck Collaboration
Participation in ClimateCheck@SDP 2025 shared task and research collaboration on efficient claim verification systems
Q4 2025: Knowledge Graph Integration
Connecting current research and developing sophisticated knowledge representation
2026: Real World Applications
Deploying expert fact checkers in production environments
Ongoing: Community & Funding
Securing funding for core contributors and building sustainable open source community
Research Team & Collaborators
Collaboration with Leading Research Institutions
Related Research
View all posts »Efficient Climate Claim Verification: Fine-Tuned BERT Models Outperform Large Language Models
A new study demonstrates that fine-tuned BERT models achieve competitive results with large language models for climate claim verification while processing claims 11-380 times faster, offering a more sustainable and practical solution for real-time fact-checking.
Automated Fact-Checking of Climate Change Claims with Large Language Models
A comprehensive overview of the Climinator tool, designed to automate the fact-checking of climate change claims using Large Language Models (LLMs).
ClimateCheck@SDP 2025: Advancing Scientific Fact-Checking of Climate Claims
The first ClimateCheck shared task at SDP 2025 challenged participants to build systems for scientific fact-checking of climate claims, with tasks focusing on abstract retrieval and claim verification against a corpus of 394,269 climate science publications.
Climate Risk Intel: Democratizing Open Climate Risk Data
Bridging the gap between complex open climate risk data and practical decision-making through accessible platforms and tools.
AI Benchmark for Sustainability Report Analysis
Building open-source tools to benchmark AI models in sustainability report analysis, with focus on detecting greenwashing and improving transparency.
Carbonara: Carbon Tracker for Software Development
Introducing Carbonara, a carbon tracking tool for sustainable software engineering.
We are open to all sorts of contributions. In particular, we seek use-cases, feedback, and datasets, and are grateful for any exchange with stakeholders.
Join Us!
Curious?