The Global Good Fund invents technology to address the most difficult challenges faced by the world’s poorest populations. It is the result of the collaboration between Bill Gates and Intellectual Ventures, stemming from a vision to address gaps in technology and science research addressing specifically problems of the poorest billion people on the planet — those globally at the base of the economic pyramid.
We focus on technologies that have the power to catalyze at-scale impacts, very much like those impacts that technology has been able to do in high-income countries. To be successful in inventing for impact, we have to have a deep understanding of the problems before attempting to solve them through technology. This has meant a commitment to “reverse innovation” where we study problems from the prospective of those in low-income countries, and try to identify the gaps that science and technology can address. When we succeed, the resulting inventions can have population-scale impact and be “catalytic” to societal change.
To adequately understand big problems, and to identify the technology gaps and the scientific breakthroughs, with the potential of being catalytic, is very complex and has required a novel process of invention.
The essentials for success
When thinking about the needs of the base of the economic pyramid, most people assume that the approach is generally to copy what works in high-income countries, and then to reduce its cost.
“Affordability” is certainly one of the factors in this kind of work. However, “appropriateness”, whether the invention will truly address the problem in the context in which it is intended to be used, and “accessibility”, or the ability for the intended users to use and scale the invention, are the other essential strategies. We consider these the essential three “As” for success in our work.
Whether for AI-based algorithms or to inform public policy, data is essential in understanding the problems and the possible impacts of technology-based innovation.
If an invention achieves these three “As” in the context of a “big” problem, it is likely to scale-up to population-scale impacts. At times, these inventions turn out to be superior solutions for everyone, regardless of income level. While initially focused on the needs of low-income populations, some of these “reverse innovation” inventions can have the effect of technological leapfrogging an entire industry.
We saw an example of this with mobile payments systems, which developed quickly to serve the needs of “unbankable” populations, and are now emerging from countries like Kenya into global solutions providing revolutionary financial services for everyone, everywhere.
In health care, as countries with more limited healthcare resources attempt to solve urgent medical problems while expanding access and quality of their health care, we are seeing tremendous opportunities to leapfrog existing models of healthcare delivery and use technology to improve healthcare outcomes, while reducing costs for everyone.
In this way Global Good tries to be the world’s quintessential reverse innovator — that is, creating technology specifically designed for the needs of the base of the economic pyramid, while enabling those technologies to ‘boomerang’ back to the developed world.
Achieving population-scale impact takes many partners across businesses, academic research institutions, governments and non-profits to make it work. The role of the private sector is essential for scaling up this kind of innovation. Novel models of private-public and philanthropy are needed to incubate, develop and ultimately scale up this type of innovation. But by leveraging the opportunity that reverse innovation provides to innovate globally while solving base-of-the-pyramid problems, there is a realistic opportunity for those companies in the private sector to disrupt and leapfrog in their market everywhere, be sustainable in emerging markets, and successful globally.
The importance of good data
Much of the future of tech for good depends on good data. Whether for AI-based algorithms or to inform public policy, data is essential in understanding the problems and the possible impacts of technology-based innovation.
As an example, Global Good made a significant investment in creating and operating the Institute for Disease Modeling, which is one of the leading epidemiological modeling groups in the world. With predictive models based on stochastic methods, we can estimate the probability of certain outcomes and weigh the likely outcome of certain innovations vs. others to optimize impact and target product profiles as well as optimize implementation. These activities are proving fundamental to select and understand
problems and identify the best possible solutions.
The power of AI: Better clinical decisions at lower cost
Some of the most transformative opportunities for medicine in our space lay at the intersection of medical diagnostics and artificial intelligence (AI).
A key gap in clinical care in low-income countries is the lack of specialists and clinical laboratory infrastructure. AI, combined with new developments in imaging, immunohistochemistry, materials science and genomics can deliver revolutionary clinical decision support systems that can make non-specialist clinicians, achieve or surpass the clinical effectiveness of the best specialists.
This is fundamentally important for low-income countries, where there are few specialists and almost no clinical laboratories. This is potentially also revolutionary for high-income countries because it can move medicine out of expensive tertiary care and into primary care, or in some cases the home, driving down costs, while improving clinical outcomes.
When innovation is approached in a multi-disciplinary way, there are significant opportunities to leapfrog to the state-of-the-art. Faced with a lack of clinical laboratories in low-income countries, we have found opportunities to transform diagnostic technologies using the latest developments in material science, genomics and imaging to provide point-of-care technologies that move the decision from central labs to the point of care. These innovations have the potential to even empower patients to monitor themselves at home, potentially changing the healthcare delivery model in some areas.
As an example, Global Good is working on AI-based ultrasound imaging, in which a deep-learning ultrasound machine can automatically detect the onslaught of pneumonia, and its progression, or response to treatment, with better predictive value than current standards of care involving X-rays and human specialist interpretation.
Other examples include AI-based automations in pathology, hematology, parasitology and microscopy as seen with the Global Good-developed EasyScan_GO microscope that we’ve introduced to the market with Chinese microscope company, Motic (see video below).
Understanding the limits of AI and Big Data
There are some warning flags, however. Big data availability can lead us to believe that simply applying AI to any data set can solve any problem. Unfortunately, in medicine, sometimes a statistical correlation does not mean causation.
It is important that we recognize that much of what we call artificial intelligence today is actually statistical intelligence, and therefore better applied to problems that can benefit from probabilistic solutions.
It is also important that data training sets and ground truths be carefully developed with clinical validations. Understanding the limitations of the technology is key in developing useful products with the right clinical safety profile and predictive value.
The future of automated quality diagnosis begins now. Global Good and Motic introduce a breakthrough AI-powered microscope to fight drug-resistant malaria.