Here’s how developing countries can reduce the Artificial Intelligence gap
Upward-facing shot of a structure. (Image Coline Beulin via Unsplash).

Here’s how developing countries can reduce the Artificial Intelligence gap

AI is essential for productive transformation and job creation in developing countries.

By Marco Kamiya

F. Scott Fitzgerald1 said that the “test of a first-rate intelligence is the ability to hold two opposed ideas in mind at the same time and still retain the ability to function”. This statement holds true with reference to today’s lopsided debate on Artificial Intelligence (AI). The negative perceptions of AI are associated with potential privacy violations and the spread of disinformation, the erosion of human rights and the replacement of labour, leading to unemployment and further inequality.

Yet AI is emerging as one of the critical technologies of the Fourth Industrial Revolution (4IR), and large corporations as well as small and medium-sized enterprises (SMEs) in developing countries could benefit considerably from using AI to improve their level of productivity. To drive the productive transformation in SMEs in developing countries, three factors need to be considered: (i) the changing nature of productive activities beyond manufacturing; (ii) the exponential growth of digitalization and AI; and (iii) the practical use and applications of AI in productive activities.

Artificial Intelligence beyond manufacturing

AI refers to the ability of a system, computer or robot to learn and carry out tasks that are usually performed by humans who have the capacity to reason and act. Typical examples of AI applications are production robots, smart assistants, self-driving cars, automated financial analyses and investments and travel booking agents, just to name a few. One key feature of such technologies is their combinatorial effects and the exponential growth they imply.

When used in manufacturing, AI contributes to cost reductions for firms, simplifies their processes, streamlines supply chains and saves resources. Moreover, it enhances predictive machine maintenance, minimizes scrap, increases defect zero yields, forecasts components demand and estimates inventories. AI is also revolutionizing customer service – chatbots can be used to provide 24-hour customer support and to predict individual customers’ needs.

Although the role of manufacturing in development has been key from a historical perspective, attention is increasingly shifting towards the potential role of services for growth in developing countries. The question therefore arises whether ‘platformization’—the combination of the production of physical goods with services and digital platforms—could accelerate productivity in developing countries (e.g. by providing local firms access to technologies and markets).

The need for speed

Technology is developing at an exponential rate and the use of AI in production activities is on the rise worldwide. China is one of the leading AI users in a range of industries, from production, and distribution to urban planning2; in Brazil, Alice Assistant, which was developed by a start-up, is used in farming and agriculture to improve harvesting processes3; in Nigeria, RxAll has developed an application to identify counterfeit drugs.

Planning for digitalization, technological upgrading and the use of AI in production or supply chain activities pose a major challenge for SMEs in developing countries45, due to the lack of adequate technological knowledge and financial resources, skills, absorption capabilities, availability of technologies as well as proper legal frameworks. Although SMEs in high-income countries face similar challenges, the use of AI is not necessarily indispensable if they occupy a niche in the market and produce highly specialized goods not offered by others, but most of the developing world need to accelerate the catching up on AI capacities (see figures below).

AI players worldwide, 2009-2018

Note: The map includes three types of players located in the area: firms, research institutes, and governmental institutions.

Source: European Commission, Joint Research Centre (JRC) (2020): AI TES Dataset 2019. European Commission, Joint Research Centre (JRC).

As AI presses forward, understanding its potential and how it can improve firm productivity is crucial. Hence, SMEs in developing countries need to be made aware of the potential of digital tools and of the availability of global and regional technologies. Failure to do so could intensify the AI divide between industrialized and developing countries where advancements based on basic research and science cannot be absorbed due to weak ecosystems and lack of effective innovation policies.

Think in potentials

AI and digital transformation can help find solutions to global problems such as climate change and environmental degradation, energy efficiency, food production and urban-rural disparities. Although some jobs will become redundant, AI will create new ones that require both decision-making skills and creativity. If used smartly, new technologies can become a source of competitiveness and contribute to a decrease in inequality and the achievement of the Sustainable Development Goals (SDGs).

One option for SMEs in developing countries is to design and implement technologies that do not require access to complex neural networks and basic science, and that can use imported advanced sensors and actuators. Examples include the implementation of a digital twinning system for remote manufacturing operations; sensors and actuators that enable virtual improvements of production systems prior to physical installation; the use of AI-forecast-fed parts and components of end-of-life vehicles to produce equipment needed locally, such as incubators for premature babies, dialysis systems, cooling equipment, etc.; applying a Geographic Information System (GIS), big data and analytics to analyse agricultural field health and improve yields in agribusiness-related activities; and designing and locally 3-D printing defective parts in production machinery rather than purchasing replacement parts in local or international markets and importing them.

AI R&D activity score, worldwide, 2009–2020

Note: R&D (Research and development), O. Asian (Other Asian countries), O. European (Other European countries), O. American (Other American countries).

Source: AI Watch Index 2021, European Commission, Joint Research Centre, 2022.

Ideas for better policy design based on practical uses of AI

Because firms and the technologies they use differ considerably depending on the region, industry and business environment they operate in as well as their income level, we must redouble our efforts to gain a better understanding of the context for digitalization and AI. Countries’ innovation ecosystems must be continuously developed and strengthened. One way to achieve this is by establishing smart manufacturing centres to train firms in digital technology applications, thereby expanding countries’ innovation ecosystems. Upgrading skills and spreading knowledge among firms in developing countries is crucial if they are to learn how to use more complex technologies such as AI for production. In Tunisia and Côte D’Ivoire, for example, training in smart manufacturing is being provided to youth with the objective of nurturing an entrepreneurial ecosystem.

New technologies can also be used to address global challenges, in particular the environmental and food security dimensions. A UNIDO project provides support to start-ups and SMEs that apply frontier information and communication technologies to contribute to a clean energy transition and to climate change mitigation through the Global Cleantech Innovation Programme (GCIP). A model to support an aquaculture transition in the Mediterranean towards sustainable and circular practices is currently being developed as well. AI enables the optimization of the FCR (Feed Conversion Ratio), a key parameter for resource-efficient aquaculture. Using underwater 360° cameras, the system analyses fish behaviour and satiety, based on which input signals are provided to automatic feeders to optimize feed rations, thereby reducing the dispersion of foreign substances in marine ecosystems. UNIDO6 also supports the design and implementation of National AI strategies, an initiative that started with the Government of Jordan and will be extended to other countries.

Although the leading industries in many developing countries, namely agriculture, mining and fishing, are considered to be unsophisticated, they could represent a potential source of growth when new technologies are implemented. In Namibia, for example, satellite imagery technologies are being used to fight invasive species and to thereby improve food security. AI can also be used by farmers as a yield predicting tool. In the mining industry, Chile’s copper giant, Codelco has upscaled its production with robotics, AI and big data, and has improved cybersecurity and procurement processes, thus scaling efficiency.

The bottom line

While the end game is uncertain, the advancement of digital transformation and AI is inevitable. What is indisputable, however, is that efforts to effectively promote innovation and national AI initiatives for industries in developing countries need to be stepped up and new technologies for productive transformation implemented to transform the digital divide into digital opportunities.

  • Marco Kamiya is United Nations Industrial Development Organization (UNIDO) Representative for Indonesia, Timor Leste, and ASEAN Affairs based in Jakarta.

Disclaimer: The views expressed in this article are those of the authors based on their experience and on prior research and do not necessarily reflect the views of UNIDO (read more).

Read next