Last Monday, 18 May, we kicked off our new AI Learning Labs initiative with an online roundtable discussion on the question: ‘How can we build a responsible climate chatbot?’.
The full recording is available with automatically generated subtitles above.
We covered a lot of ground in a wide-ranging conversation over two hours. While many of the reflections were specific to questions of climate, many also applied to responsible design, development and use of generative AI in general.
Even as we explore the use of chatbots for knowledge delivery, how do we avoid falling into the trap of thinking of AI itself as a knowledge base? We discussed how differently individual LLMs answer questions to the same questions depending on language and context—and the limitations of testing outputs for misinformation when every user sees something unique.
We spoke about the importance of sourcing data, especially from the Global South, and the approaches different groups are taking to fill the gaps in mainstream media and science. For instance, The Climate Academy looks at climate from a position of equity, and takes issue with mainstream public discourse on climate that simply looks at territorial emissions per country. How could a chatbot deliver responses that align with this worldview?
Finally, we touched on the emissions of LLMs and the practicality of mid-size and small-size models for most daily needs. We heard about alternative hosting options, transparent service providers, and were urged to question whether a chatbot is even the best approach for using AI to impart knowledge. For sure, both The Climate Academy and #semanticClimate said they imagine their prototype as a community tool rather than a solution on its own.
A heartfelt thank you to the dozens of people who contributed tirelessly in the chat and to our guest speakers:
- Gitanjali Yadav, #semanticClimate
- Robert Keus, GreenPT
- Ihor Samokhodsky, Policy Genome on disinformation in AI across models
- Adrianna Baravalle, Director of the Exponential Technologies Laboratory at the University of Austral in Argentina
- Matthew Pye, founder of The Climate Academy
If you are interested in our continually evolving learning exchanges on this topic, drop into the Open Knowledge Forum, where you can share links or reflections. Also: #semanticClimate welcomes contributions to the prototype they are building.
About
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 conversation:
- Open Knowledge Forum
- School of Data community
- Contact our team: 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.
