Let's build together a world where all knowledge is
accessible to everyone
Get in touch
accessible to everyone
Making climate knowledge accessible to everyone, with accuracy, transparency, and trust at its core
Meet the Climate Academy Chatbot (CABot), a responsible climate chatbot for trusted, transparent climate information.
CABot is a responsible, climate-friendly AI chatbot that can answer questions about climate change using pre-approved literature, datasets, and conversational guardrails. The goal is to make climate knowledge more accessible to everyone, keeping information clear, factual, and credible. The chatbot is grounded in trusted sources, clear design choices, and thorough testing, so that it can answer questions appropriately.
Unlike mainstream AI chatbots that may generate distortions and misrepresentations about climate change, CABot is built to provide answers that are traceable, transparent, and trustworthy — exactly what climate educators and learners need.
CABot is the first public output of Open Knowledge's AI Learning Labs, the result of a collaborative effort by The Climate Academy and #semanticClimate.
What makes CABot a responsible chatbot
|
|
Reliable Knowledge Answers come exclusively from sources that have been selected and curated: The Climate Academy book and a climate encyclopedia built by semanticClimate using material from the Intergovernmental Panel on Climate Change (IPCC/AR6) Working Groups. |
|
|
Grounded Answers CABot generates responses strictly from retrieved content. If the chatbot doesn't have the information, it will say "I don't know" rather than inventing an answer. |
|
|
Source Citations Every answer includes a reference to the original source, allowing users to verify credibility and dig deeper. |
|
|
Fairness The chatbot expresses how climate change affects different communities from different geographical regions in a balanced and respectful way. |
|
|
Accessibility The chatbot uses simple language to present climate knowledge in such a way that everyone understands. |
How CABot works
The chatbot transforms raw HTML content into a searchable semantic knowledge base, then answers user queries in real time with verified citations.
The system operates in two phases: a one-time ingestion pipeline that builds the knowledge base, and a real-time query loop that delivers cited answers in seconds. Embeddings are generated once during ingestion. At query time, only semantic retrieval and LLM generation run – keeping responses fast.
Related content
 |
Open Knowledge’s AI Learning Labs is an initiative that aims to experiment with AI, translate knowledge from social sector organisations around the world, and produce public, multilingual AI-literacy resources tailored for organisations addressing similar issues elsewhere.
Together, we will catalyse learning and develop replicable methods to help organisations build AI skills, use AI responsibly, and develop their own AI projects. All resources will be openly available at School of Data.
Pilot Partners
Climate Academy Chatbot (CABot) was developed in the first half of 2026 during the Responsible Climate Chatbot Lab from a partnership by the following organisations:
Get involved
We welcome contributions from experts and practitioners from anywhere who are interested in learning about and replicating our methods, as well as adapting our resources to their own languages and cultures.
-
Join the discussion on the Open Knowledge Forum
-
Give feedback on CABot
-
Explore the code on GitHub
-
Join the School of Data community here
In addition to the channels above, you can also contact us at info@okfn.org.
This project has been made possible thanks to the generous support of the Patrick J. McGovern Foundation (PJMF). We are grateful for our ongoing partnership in promoting digital literacy and investing in AI for the public good.
Learn more about its funding programmes here.
Don't miss a thing
Stay on top of what's happening in the #OpenMovement around the world. (We will never spam you!)