How much can AI be a protagonist in the development of countries in the Global South and how much can it be antagonistic, going to increase the gap between countries with different levels of development?
The risk is real and present. AI generally requires the Internet and large computing resources.
As with other technologies, the opportunities offered in terms of access to information, are exponential. At the same time, as has been seen in the past, these technologies are then effectively controlled, both in the development and in the use phase, by a few parties who are able to concentrate the resources (of infrastructure and knowledge) needed to manage them. In the face of great opportunities, great risks grow accordingly. These risks can be mitigated essentially by two factors: the first is the use of open technologies and the second is an innovative role for policy to regulate the use of these tools for the collective good.
What are the most relevant aspects of AI as applied to the world and professions of development cooperation?
It was immediately clear to everyone how Chat GPT can be a faithful companion in writing. But the possibilities are much broader and extend to all NGO work. I think the first area to apply AI to is project-design. AI can be applied at different stages of this process, whether at a preliminary stage through analysis of calls for proposals and identification of a project, as in defining project logic. It can also help within the body of the initiative, with the definition of monitoring plans and identification of indicators, creation of survey tools and progress reports, writing internal and external communications, managing human resources, and defining and creating social communication plans. Finally, AI can serve for the creation of training content and automatic translations of all content produced in any language and much more.
Are there any protocols, guidelines, international or national programs that point to a virtuous path for AI development in countries of the Global South and/or any already active projects you would like to mention?
Regarding guidelines, I would like to suggest this document published by USAID (United States Agency for International Development) entitled “ARTIFICIAL INTELLIGENCE (AI) ETHICS GUIDE” within which the different facets of artificial intelligence are addressed. The guide reports information regarding the challenges, opportunities and risks of AI by presenting concrete examples of discussion on the ethics of using this tool and how, they should be resolved.
Then I would also recommend an interesting academic research paper on natural language processing (NLP) in the humanitarian sector. In the paper, it recognizes how modern NLP tools have the potential to support humanitarian action at different stages of the humanitarian response cycle, however, lack of awareness about the concrete opportunities offered by the most advanced techniques and constraints posed by resource scarcity limit the adoption of NLP tools in this field. Truly a fine text to explore a topic still in its infancy.
If you had the opportunity to develop two direct interventions/policies in this field (without budget constraints) what would you activate?
First would be the creation of open source, context-sensitive LLMs (Large Language Models) with the collaboration of universities and practitioners from vulnerable backgrounds.
The second would be training teachers and practitioners in countries of the Global South in the use of AI.
Can you recommend a book or movie to continue to explore this topic?
There are several books and movies that talk about this theme but, personally, I would suggest as readings: “I Robot” by Isaac Asimov, which is from 1950 but still contemporary. Then, I would add “Machines Like Me” by Ian McEwan.
As for movies I would recommend “Her” by Spike Jonze starring Oscar winner Joaquin Phoenix and “E noi come stronzi rimanemmo a guardare” Italian film directed by Pif starring Fabio de Luigi.
NB: The interviews reported here are not part of paid commercial services. They are for the sole purpose of sharing ideas, projects and reflections among De-LAB newsletter subscribers.a gli iscritti alla newsletter De-LAB.
