of Gates Foundation and OpenAI On Wednesday, a $50 million “experiment” was launched to “advance AI capabilities for health” in Africa. Horizon 1000 has committed “finance, technology, and technical support” to deploy AI solutions to 1,000 primary healthcare clinics in Africa by 2028.
“AI will be a scientific marvel no matter what, but for it to be a social marvel we need to figure out how to harness this amazing technology to improve people’s lives,” said Sam Altman, CEO of OpenAI. media release.
“The goal is [health care] “It will be much higher quality and hopefully twice as efficient as it is today, removing paperwork and organizing resources so patients know what’s available and when to come for an appointment,” said Bill Gates, CEO of the Gates Foundation. Session at the World Economic Forum (WEF) Wednesday in Davos.
Starting with Rwanda
The pilot will begin in Rwanda and then expand to Kenya, South Africa and Nigeria, Gates added.
Rwanda is already exploring the use of AI to help health workers diagnose diseases, free them from administrative burdens, and model the course of the disease.
Paula Ingabire, Rwanda’s Minister of Information, Communication, Technology and Innovation, told WEF that her country has been working on technological solutions to “pain points” for more than 20 years.
For starters, the country has rolled out internet access to about 97% of its population. This is a significant achievement in a country where most people live in rural areas.
The company is currently building “some of the foundational digital infrastructure that will enable and empower us.” [technological] It’s progress,” Ingabire said.
One of Rwanda’s goals is to use AI to create “decision support tools” for the more than 60,000 community health workers (CHWs) who provide primary health care to communities across the country.
With malaria accounting for about 70% of the cases CHWs handle each year, CHWs want AI tools that can improve diagnosis and better predict when and where malaria cases will occur, Ingabire said.
Rwanda is already using a combination of drones and AI to combat malaria, with drones identifying and spraying mosquito breeding areas and AI helping predict and model the disease.
Two years ago, Rwanda set a goal of quadrupling the number of health workers in four years, a goal that has already been largely achieved.
“But they’re going to need these tools to support the delivery of better care. Some of the administrative tasks they’re working on can be done using AI. So they’re focused on providing better, more targeted care to the population,” Ingabire said.
The government also wants to use AI to improve demand forecasting for health products to prevent drug stockouts.
“We started this digital transformation journey over 15 years ago. We have a lot of data that we’re not using. It’s important that we build a national data intelligence platform that helps us. Once we build these models, they need to be trained on our own data, they need to be context-specific, and they need to be deployed to address real-world problems.”
Ingabire added that her country is also in talks with Anthropic, an AI company that developed the large-scale language model Claude, and is “looking at how we can deliver an instant health intelligence platform that feeds the entire national health planning system and allows us to better allocate resources.”
AI-based tuberculosis screening
Peter Sands, CEO of the Global Fund to Fight AIDS, Tuberculosis and Malaria, told WEF that the foundation has invested $170 million in AI-based tuberculosis testing over the past four years.
This is one of the largest single applications of AI and health and has a “very significant impact,” he added.
One example of how the foundation has used AI-based tuberculosis testing is in refugee camps.
“There are well over one million Sudanese refugees in Chad. We are working with the Chadian government to set up mobile clinics that go into these refugee camps and test for tuberculosis,” Sands said.
Because there were no radiologists available, “if you want someone to read your screening, there’s no other option.” [but AI]” he added.
However, Sands cautioned that some “very fundamental issues” still need to be fixed to enable the deployment of AI-driven solutions. Primarily, many primary health care facilities in Africa lack internet connectivity, and some even lack electricity.
He also argues that the use of AI should be “framed around problems that need solutions, rather than a bunch of tools that need problems to solve,” similar to how people “run around with a bunch of hammers looking for nails.”
He added that developing tools is easier than finding “someone who can actually use them to make things happen.”
Are LMICs progressing faster than rich countries?
Gates believes that the deployment of AI in health care could be faster in the world’s developing countries than in the “rich world” because “the need is so great that governments are embracing this and making sure they are moving full speed ahead.”
“The $50 million commitment is just the beginning. I believe Africans should have this ‘health advisor’ without having to pay anything. It should be a basic feature available to them.”
“When you go into the health care system, instead of filling out forms and re-explaining everything, the AI you’re talking about will summarize it for you… eliminating the paperwork all together.”
Sands also believes that low- and middle-income countries may be able to adopt AI tools faster than developed countries, where regulations are tighter and AI is more likely to eliminate jobs.
“One of the reasons why this movement may start sooner in middle-income countries is because there will be less resistance from people who say, ‘This took my job away,’ or ‘I don’t want to change my ways,’ because it makes up for the fact that those people don’t exist.”
Image credit: Cecille Joan Avila / Partners In Health.
Combating the infodemic in health information and supporting health policy reporting from the Global South. Our growing network of journalists in Africa, Asia, Geneva and New York connects the dots between local realities and global debates with evidence-based, open-access news and analysis. To contribute as an individual or organization, click here.
