Open Knowledge is collaborating with governments to prototype an integration with open data portals that enables users to ask natural-language questions and receive responses traceable directly to the underlying datasets.

Credits: Kuu Akura, Douglas Lopez/Unsplash

The problem

  • As AI becomes a primary gateway to information, public institutions that publish open data face a new and complex challenge: how can they ensure that AI tools referencing their datasets generate accurate, transparent, and verifiable answers — given that existing mechanisms for transparency, provenance, and verification are not yet adapted to this AI-driven information landscape.

The opportunity

  • We are building a direct connection — using a Model Context Protocol (MCP) that connects AI directly to official data portals such as CKAN. With this, we can enable AI-generated responses to be traced back to their original datasets. By partnering with national governments to test this on real-world open data, we can create a proven, repeatable model for responsible and transparent AI use with public data.


 

This project is a collaborative initiative grounded in real-world challenges. Two pilot programmes are taking place in 2026 under the umbrella of the Digital Public Goods Alliance (DPGA) in partnership with national governments to become a reproducible and scalable reference of responsible and transparent AI integration with open data infrastructure.

Government of Brazil
Government of Uruguay

 

Follow our progress, from research notes to community stories:

4 May 2026
Brazil and OKFN partner to enable citizens to ‘talk’ to national data 4 May 2026
7 April 2026
'An Honest Reflection on the Integration of LLMs into Open Data Portals', by Patricio Del Boca 7 April 2026
25 March 2026
[Recording] Technical community consultation on MCPs and LLMs 25 March 2026
11 March 2026
Roundtable: 'CKAN at 20 and the Future of Open Data' 11 March 2026
16 December 2025
Announcement of Strategic Funding to Enhance Multilingual, Sector-specific AI Literacy and Develop Trustworthy AI for Open Data 16 December 2025

 

 

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.
 

↗
Join the discussion on the Open Knowledge Forum
↗
Join the School of Data community

 

CKAN is the world’s leading open source data management system. Created at OKFN two decades ago and a recognised digital public good (DPG) since 2023, CKAN is the backbone of open data infrastructure on every continent — powering national governments, intergovernmental bodies, and research institutions. 

🎉 We are celebrating 20 years of CKAN with various activities throughout 2026.


 

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.

 

colors
Let's build together a world where all knowledge is
accessible to everyone
Get in touch

Don't miss a thing

Stay on top of what's happening in the #OpenMovement around the world. (We will never spam you!)