India's AI strategy in perspective

M Rokonuzzaman | Published: December 21, 2018 21:05:07 | Updated: December 22, 2018 21:16:59

Every major country in the world has been responding to the potential of Artificial Intelligence (AI) technology. Unlike many other technologies, AI has the potential of causing massive harm as well as good to virtually every country as well as firms. On one hand, AI poses a threat of massive job loss in labour-intensive productive activities, and on the other, it has the potential of increasing efficiency and effectiveness of productive activities, while creating high paying jobs for innovation and reducing pollution in a profitable manner.  To minimise negative implications and maximise gain from AI, major countries are in the race to come up with smart strategies and policies. To respond to AI, in June 2018, India released a discussion paper on AI--"National Strategy for Artificial Intelligence."

Countries that have released AI strategies and policies well ahead of India are USA, China, Japan, UK and France. Lately, Germany has also revealed its 3 billion Euro plan to become AI powerhouse. Neither all these countries are focusing on the same application areas, nor are they adopting the same strategy and policy responses. For example, the USA has been focusing on industrial productivity, homeland security and defence. On the other hand, China has been pursuing the fusion of defence and civilian applications to harness the power of AI. For India, AI poses threat to massive factory job loss. But, for China, AI is a blessing for killing jobs with machines in dealing with aging factory workforce due to One Child policy. Despite job loss threat, AI could be a blessing for India to pursue cleaner production techniques and increase agricultural yield by reducing wastage of farm inputs like water and fertilizers.  

By the way, AI is not a new phenomenon, with much of its theoretical and technological underpinning developed over the past 70 years. AI is a constellation of technologies that enable machines to act with higher levels of intelligence and emulate the human capabilities of sensing and comprehending. But despite the theoretical development and demonstrations of potentials in research laboratories, application of AI in productive activities has been very limited. Both the lack of technological feasibility and financial non-viability created the barrier on the way of adding human-like intelligence to machines.  But the recent growth of smartphone has made vital components of AI highly capable for profitable deployment of AI solutions even within the context of developing economies like India. For example, AI fitted irrigation solutions can ensure optimum watering to crops in maximising yield, while reducing water wastage by bringing down India's staggering water footprint from over 2,000 cubic meter per ton of rice to global average of 1,325 cubic meter per ton. Such AI-augmented optimum watering can also help India reduce the cereal yield gap between USA's over 8,000 kg/hectare to India's less than 3,000 kg/hectare.

According to a report of an international consultancy, AI could potentially deliver an additional economic output of around $13 trillion by 2030, boosting global GDP by about 1.2 per cent a year. To capitalise on this potential, India has targeted 5 key areas such as i. healthcare, ii. agriculture, iii. education, iv. smart cities and infrastructure, and v. smart mobility and transportation. In the strategy paper, healthcare attempts to increase access and affordability of quality healthcare by leveraging AI. Through AI, India would also like to improve access and quality of education, address constraints facing cities and infrastructure making them efficient in offering connectivity for the burgeoning urban population, and also offer smarter and safer modes of transportation and better traffic and congestion problems.

India has been pursuing AI strategy to address issues affecting all, terming it AI for all. On the other hand, China has been looking into AI as a strategy to attain supremacy in industrial economy and defence. In the AI strategy, although India has overlooked industry, for China AI has been a survival strategy for addressing productivity and growing wages. It's understandable that job loss due to AI is a sensitive issue to India. As reported by BBC, India's transport minister vowed to ban self-driving vehicles to protect jobs for millions of drivers. On the other hand, China has been desperate to deploy intelligent machines and advanced robotics to kill 100 million jobs over next 10 to 15 years to address a shortage of factory workers caused by the retirement of an aging workforce. Such a strategy is helping China to slow down the migration of factories to labour-surplus countries and protect high paying jobs for the white collar workforce.

The implementation of AI strategy to derive profitable return from the investment is also another challenge in this globally connected economy. Despite success in IT service export, India's institutional capacity in carrying out AI research leading to commercial innovations appears to be quite weak. In both publications and patents, India has a fragile track record in comparison to China and many other advanced countries. It's going to take almost 10 to 20 years to turn AI potential in target areas into profitable innovations. How to finance this long journey is an important issue. Who will be the customers for the initial emergence of premature AI innovations? Historically, it has been found that the military played an important role to fill the financing gap, as often premature solutions served the purpose of the military.  For example, missile defence system with 80 per cent success rate is good enough for the purpose of the military, as opposed to having no defence at all. But for the purpose of self-driving vehicles, the success rate of detecting obstacles must reach almost 100 per cent to qualify for the deployment in city streets. Coupling of AI strategy with defence and homeland security may be an essential requirement to succeed in commercial innovations. Both the USA and China have clearly stated fusion of defence and civilian applications to leverage AI potentials.

Due to the lack of focus on industry in approaching AI, India's industry runs the risk of losing competitiveness due to the rising use of AI by competing countries like China or Germany. On the other hand, due to delinking of defence, India's AI strategy runs the risk of suffering from lack of customers for initial versions of solutions in the long journey of taking AI ideas to civilian market.

Like India, all developing countries need to fuse aspiration with speed in leveraging AI innovation in this globally connected economy, so that the benefit from AI outweighs the likely erosion of competitiveness due to the labour substitution effect of AI. 

M Rokonuzzaman Ph.D is academic and researcher on technology, innovation and policy.

Share if you like