The Accenture Institute for High Performance recently released research revealing that, by 2035, artificial intelligence (AI) could double annual economic growth rates in developed economies.
The study compared economic output in each country in 2035 under a baseline scenario based on current assumptions against one showing expected growth once the impact of AI has been absorbed into the economy. In the US, the annual growth rate went up from 2.6% to 4.6% - an additional $8.3 trillion in gross value added (GVA) with widespread AI adoption included. In the UK, AI could add an additional USD $814 billion to the economy, increasing the annual growth rate of GVA from 2.5% to 3.9%.
However, it is developing economies where AI is likely to have the more significant impact. We have already entered a period in which enormous technology-driven change is helping to address a number of challenges in developing economies. AI technology, in particular, has extremely strong developmental implications. There are still a number of challenges towards implementation of such technology. The infrastructure, for one, is not necessarily capable of integrating all AI technologies, so it is not a case of inventing something new and then dropping it into a developing economy. Yet, while the first priority must be to build infrastructure - next-generation telecoms, power and agriculture systems - so that AI can be used, there are already a number of ways that it can be applied.
There are three main areas that AI can be of benefit.
There are two pressing concerns for the majority of people in developing countries: access to water and food. To provide citizens with food, smallholder farms must be able to produce enough. However, currently, research infrastructure and agricultural extension systems capable of supporting smallholder farmers are sadly lacking. AI is capable of increasing the yield of farmland under tillage in developing countries, with machine learning algorithms used in drone technology to both plant and fertilize seeds at a speed beyond human abilities.
Another application of AI for food management in developing economies is the identification of disease in crops so they can be more easily treated. A team of researchers at Penn State and the Swiss Federal Institute of Technology (EPFL) have fed a network of computers with over 53,000 photos of both healthy and unhealthy plants in an attempt to recognize specific plant diseases. Such technology will provide the basis for field-based crop-disease identification using smartphones. The system has been able to identify both crops and diseases – from photos – with an accuracy rate of up to 99.35%.
For NGOs and charities, determining where resources are needed is vital to helping those most in need. If available resources are not properly utilized, the scarcity makes a bigger dent. This is another area where AI can help greatly. It can be used to learn to analyze multiple factors at the same time in a way that humans cannot which can show, say, where a drought could occur, how many people it is likely to impact, and what is required to fix the problem.
For example, ’Harvesting’ is a startup using machine learning to analyze satellite data of the Earth’s surface. They are trying to pinpoint areas in need of investment in the water and tools needed for farming to help institutions distribute money more efficiently.
CEO of machine learning startup Harvesting, Ruchit Garg, recently noted of AI that, ‘Our hope is that in using this technology we would be able to segregate such farmers and villages and have banks or governments move dollars to the right set of people.’
The Ebola virus wreaked havoc on African communities, as numerous outbreaks have over the years. In the case of Ebola, Barbara Han, a disease ecologist at the Cary Institute of Ecosystem Studies, said, ‘Using machine learning methods developed for artificial intelligence, we were able to bring together data from ecology, biogeography, and public health to identify bat species with a high probability of harboring Ebola and other filoviruses. Understanding which species carry these viruses, and where they are located, is essential to preventing future spillovers.’
The main advantage of Machine Learning is its ability to deal with complexities. With a number of variables interacting at one time, findings can become difficult to interpret. Machine Learning side steps this. On this issue, Han says: ‘The algorithm doesn’t care how the variables are interacting; its only goal is to maximize predictive performance. Then we human scientists can step up.’
Machine learning technology is the most effective way not only when it comes to understanding the spread of disease, but also providing relief. We are looking at future where machine learning could feasibly identify a disease, develop a cure, locate where the outbreak is likely to strike next, and then transport the cure there in autonomous vehicles, all with minimal human interaction.
There are many bridges to cross before this becomes a reality. However, while AI in developed countries will have a major impact, it could be necessary for their very survival, and it is vital that everything is done to ensure the infrastructure is in place to take advantage of every advancement in the technology.