Background 

Artificial Intelligence (AI) has significant potential to transform healthcare around the world. But as AI technology continues to advance, there is an urgent need to position low- and middle-income countries (LMICs) to lead the design and co-creation of AI-enabled technologies – thereby helping to improve the health and well-being of women, children, and vulnerable communities.  Through this call for proposal, we are emphasizing AI solutions that are locally driven and owned hence more relevant to address the needs of people they intend to serve and more likely to be accepted and used by local communities.  

Responsible global use of AI entails a safe, equitable, transparent, reliable, and beneficial process that is adhered to with a high level of accountability. As the world rapidly moves to seize AI's opportunities, it is imperative to monitor and mitigate the safety, ethical, equity, and reliability dimensions of AI deployment. This will allow the enormous resilience, creativity, and commitment of researchers, scientists, and policymakers to capture the full capability of AI for lasting good. 

Grand Challenges (GC) partners including GC Brazil, GC Ethiopia, GC India, GC Senegal, GC South Africa, and GC Africa (pan-African), along with the Bill & Melinda Gates Foundation, the Patrick J. McGovern Foundation, and the Pasteur Network, have jointly discussed the need for an equitable and responsible approach to the use of AI and specifically Large Language Models (LLMs) in LMICs. This call by the various GC partners is a follow-up to the initial call by the Bill & Melinda Gates Foundation representing collaborative steps towards identifying, nurturing, and catalyzing the creativity, energy, and skills that researchers, implementers, governments, and technical partners have demonstrated in solving specific challenges in their countries and regions through LLMs. We are optimistic that this family of related AI RFPs will provide a platform for potentially transformative solutions to improve the lives and conditions of numerous vulnerable communities. 

Note that this request for proposals is being launched by six Grand Challenges partners: African researchers located in Ethiopia, Senegal, and South Africa, should apply through the corresponding country-specific Grand Challenges portal; African researchers located in all other African countries should apply through the Grand Challenges Africa portal; Brazilian researchers located in Brazilian organisations should apply through the Grand Challenges Brazil portal; and Indian researchers located in Indian organizations should apply through the Grand Challenges India portal.

Eligible Pasteur Network sites located in Africa should apply through Grand Challenges Africa and Pasteur Network sites located in Brazil should apply through Grand Challenges Brazil. All other eligible Pasteur Network sites (not located in Africa or Brazil) should apply through Grand Challenges Senegal. Please consult the information provided in the subsection referencing geography of interest on the linked partner site to ensure that your application fits the requirements of the RFP you are applying to. 

The Challenge

This RFP seeks innovative approaches to the use of LLMs (such as ChatGPT-4, or other credible sources with equivalent capability) to advance global public health. Given the numerous open and non-open-source AI tools, we encourage/expect the applicant to select the tool most appropriate to their use case and context. 

We are looking for projects that seek to solve problems in the following areas: 

Clinical Decision Support  

• Tools for that can be used by frontline health workers or clinicians for improving the diagnosis and management of health conditions and/or healthcare delivery 
• Support for health guidelines refinement and adherence  
• Diagnostics interpretation, reducing costs, and overcoming at-a-distance barriers. 

Population Health and Policymaking 

• Support for policymakers in leveraging new and timely insights through routinely available, underutilized, or unused data sources of text and voice 
• Reduce time delays to transition new evidence into policy and implementation, as well as optimise resource allocation 
• Approaches that can distil information and make timely recommendations from complex and evolving datasets (e.g., forecasting disease epidemics and progression). 

Support for Frontline Health Workers 

• Personalized coaching for semi-skilled FLWs that is tailored, highly relevant, and leads to an improved quality of service and/or lower costs 
• LLM-supported solutions for workflow management (e.g., writing discharge summaries, etc.) 
• Use of LLMs to support skilled FLWs in delivering higher quality services and improved efficiency. 

Health Communications and Patient Journeys 

• Develop impactful and targeted communications tools that bridge language and literacy gaps when communicating health-related knowledge, messaging, and advice (i.e., translating from local dialects, text-to-voice, etc.) 
• Provide timely, trusted, and tailored advice to end users who are marginalized and in so doing overcome significant cultural, access and stigma concerns 
• Support patients in understanding and managing their own health status and care regimen. 

Health Systems Strengthening 

• Use of LLMs to improve interoperability of health data, systems, and programs. 

Ideally, the proposal will: 

• Work on a specific problem that has been identified as a priority in Africa. 
• Leverage AI to increase productivity and efficiency in healthcare. 
• Contribute to the body of evidence related to AI use in LMICs across global health. 
• Foster innovation with AI in the interest of supporting public health decision-makers and affected communities. 
• Place a strong emphasis on systematically observing, validating, and quantifying the improved outcomes balanced with cost-effectiveness from AI use. 
• Conform with global principles of AI use i) to do no harm ii) to leverage technology against the toughest/most relevant global health problem iii) ensures that projects are led from LMICs even though there may be a high-income country (HIC) partner iv) local voices and ideas are captured throughout the process, and v) there is a dissemination plan of the outcomes of the projects. 
• Give local communities an opportunity to provide their own perspective and cultural context so that they can decide on both i) their own thresholds for safe usage and ii) the overall utility of AI within their own lives.

Priority will be given to proposals that have: 

• Projects that have lessons/tools that can be transferred to other use cases/situations/ contexts with minimal change 
• Emphasise the importance of community-specific, culturally appropriate, and representative AI. 
• An explicit request for an AI-supported project in Africa 
• Those projects that have already completed a pilot before this Grand Challenges call, and/or 

Funding Levels, Period of Performance for Grant and Geography of Interest 

The funding level is a grant of up to USD $100,000.00, for each project, provided to the organization, with a term of up to 12 months. ii The geography of interest for this proposal is Africa, and the project needs to be led by investigators in this geography. Global partners may be included, but proposals must demonstrate at least 80% of the funding is going to an organization within the geography of interest. Application budgets should be commensurate with the scope of work proposed. 

What we are looking for: 

• Proposals that demonstrate clear applications of LLMs (such as GPT-4, Laude, LLaMA, or other credible sources with equivalent capabilities), to address a specific challenge related to global health in Africa. 
• Proposals that demonstrate the grantee has a firm grasp of the application, has done some of the stakeholder mappings and has an engagement plan with local decision-makers to ensure the proposal is successful.  
• Proposals that present a high leverage and scalability opportunity. 
• Proposals that outline a clear, feasible and reproducible methodology.
• Do not provide any thought towards scale-up and sustainability. 
• Proposals that have timely access to data, decision-makers time and interest in using AI 
• Proposals that articulate how the project will lead to impact in the near term and how those benefits will be sustained past the lifetime of the project. 
• Proposals that are driven by a shared commitment to open science, data sharing, and building collaboration and analysis infrastructure to enable discoveries that will benefit people everywhere. 
• We particularly encourage applications from women-led organizations and applications involving projects led by women. 

We will not consider funding for proposals that: 

• Do not explicitly use/reference the use of LLMs in their project execution.
• Do not have timely access to necessary data, decision-maker time, commitment, and interest (a commendation/request letter will be advantageous). 
• Do not demonstrate that most of the work proposed will be led by investigators and teams based in Africa.
• Do not include a validation plan. 
• Do not provide any thought towards scale-up and sustainability.