How India is carving out a niche for itself in the field of Artificial Intelligence

When 62-year-old computer scientist P Anandan started last September as CEO at the Wadhwani Institute of Artificial Intelligence (WIAI), he might have been apprehensive. He was signing up to work on AI in India after three decades at major global corporations and academia in the US and at home, including top teaching and research roles at Yale University, Adobe and Microsoft.
India isn’t exactly a hub for cutting-edge research in Artificial Intelligence (AI). Misconceptions abound about big data, machine learning, automation and other AI-related technologies in the context of job losses for humans. And unlike other major economies, India hadn’t yet spelled out its vision for a future with AI.

But when Anandan set out to work, he found support all around. In February, the Prime Minister inaugurated the institute. Maharashtra chief minister Devendra Fadnavis and NITI Aayog CEO Amitabh Kant were in attendance. Modi also invited Anandan to Delhi and met with him there.

“Right from the top, the government support has been fabulous. They have almost been like our working partner,” Anandan told ET Magazine. Anandan’s institute, funded by the entrepreneurs Romesh and Sunil Wadhwani, is among a string of emerging private, non-profit and government efforts to harness the power of AI to solve societal problems in India.

The focus is on areas such as agriculture, healthcare, education and infrastructure. There are also moves to channel global AI talent and resources to develop solutions that can benefit millions at the grassroot. And this month, India made its first steps in articulating what it wants to be in an AI-centric future.

A National Strategy
“Globally, no one is doing AI innovation for the social sector. India can lead here,” Anandan says. That’s indeed the overarching vision in the first major blueprint on AI that was released this month—a discussion paper from government think-tank NITI Aayog, titled National Strategy for Artificial Intelligence.

It pitches India as the AI “garage for the emerging and developing economies”. The focus will be on five key sectors— healthcare, agriculture, education, smart cities & infrastructure and smart mobility & transportation. The priorities are hard to argue with. India’s agriculture is notoriously inefficient, employing just under 50% of the population but contributing under 18% of the country’s gross domestic product.


In the US, for instance, less than 2% of the workforce is dependent on agriculture. Access to healthcare is poor—India’s life expectancy (68 in 2015) is among the lowest for BRICS nations, and so is its hospital beds per thousand people (0.9), a KPMG report said last year.

“AI is like the new electricity. The new factor of production. Like the industrial revolution, it will transform every sector. We want to exploit it to solve big socio-economic challenges that India faces,” said Amitabh Kant, CEO, NITI Aayog. This view of AI as a transformative force underpins the report as well.

While AI is commonly understood as a piece of esoteric high technology that could get too powerful for our own good, it’s really a suite of technologies like machine learning, pattern recognition, big data, neural networks and self improving algorithms, many of which have been around for a while and are now maturing.





What has changed in recent years in the promise of AI is the coming together of very advanced computing power with highly sophisticated algorithms and networks that collaborate to recognize patterns, discern meaning from vast sets of data and train themselves to get better at these tasks.

For the enthusiasts, the big question in AI might be when we will achieve machine super-intelligence, or Singularity, the point at which machine intelligence explodes to vastly surpass human intelligence. It is chillingly described as the last invention humans will make (not necessarily because it will destroy mankind, but because all the inventions thereon will be made by machines).

But for a country at India’s level of socio-economic development, the suite of AI technologies can be applied effectively to relatively prosaic concerns. And it’s already happening.

In Andhra Pradesh for instance, Microsoft is working with the state government to predict dropout rates in government schools. With granular knowledge of past outcomes across factors like gender, socio-economic background, academic performance, school infrastructure and teacher skills, the machine is able to predict with some margin of error, who among the current cohort are likely to drop out and when. This enables early intervention.

Microsoft and the International Crops Research Institute for the Semi-Arid Tropics have together developed a sowing app that uses AI technologies. It sends advisories to participating farmers on the optimal date of sowing. IBM and NITI Aayog have developed an AI-based crop yield prediction model. It’s able to offer real-time advisories to farmers on crop yield, pest outbreak, and so on.

Efforts are underway to make cancer screening and diagnosis far more accessible than it is currently, using AI technologies. Traffic, crowd management, avoiding of accidents, improvement of public facilities like parks and open spaces—are all problems with smart AI-based solutions.

Collaborative Efforts
Anandan of the Wadhwani Institute says collaboration between companies, government, non-profits and universities is key to effectively harnessing AI for India’s problems.

“The government and the NGOs do not have the capacity to tackle problems through technical innovations. We will be the connector that can help build public data ecosystems, develop solutions and help deliver at scale,” he says.

He is now helping Mumbai University rollout a master’s program in data sciences. Being a not-for-profit with an open innovation platform, close to 200 top AI scientists from universities such as MIT and Stanford have volunteered to collaborate to solve big societal problems.



The NITI Aayog report proposes an umbrella organisation to shape and implement India’s AI policies—from stitching global partnerships to picking specialised teams to pursue audacious moonshot projects. It identifies five focus areas (see box) and a two-tiered institutional structure — government-led CORE or Centre of Research Excellence to focus on core research and private sector-led ICTAI or International Centers of Transformational AI to focus on application-based AI research.

For AI to work, having vast amount of structured clean data is critical. To build data ecosystems, the report proposes a National AI Marketplace (NAIM) that will collect and annotate data and evolve deployable models. To tackle AI talent shortage, it suggests a slew of initiatives like reskilling workforce, modular certification courses and thrust on research with PhD fellowships.

The thrust of the report has been lauded by AI experts who are also familiar with India’s problems. “The report looks pretty comprehensive. I am impressed with what they have managed to put together,” says US-based scientist Subbarao Kambhampati, president, Association for the Advancement of Artificial Intelligence.

Akilur Rahman, chief technology officer at Swedish-Swiss tech giant ABB, says it tackles the topic well and in a timely manner. NVIDIA South Asia MD Vishal Dhuppar, who is passionate about AI, says the report’s thrust on public-private collaboration is the way to go. Prof V Kamakoti of IIT-Chennai, who authored a report on AI for the ministry of commerce and industry, said he found the Niti Aayog report to be “well-articulated and action oriented”.



However, sceptics abound. Many doubt government’s execution capabilities. The poor state of university research is a big worry. India lags far behind China and the US in the number of AI experts, researchers, patents filed, research published, AI startups and funding. They point out India neither has the private sector tech giants like Google and Facebook to give heft to its AI moves nor China’s financial might and state determination to make things happen.

To the naysayers, here’s what Kant has to say: “We are committed to having this strategy executed. Every major change faces hurdles and resistance. Change is the way to progress. The execution will be implemented using a METRO approach (Measurable, Time-bound, Real Outcomes).”

Gathering Momentum
With deep economic and security implications, the race for supremacy in AI is the space war of the digital age. All advanced nations are vying for supremacy, while some are focusing on niches. The US, with top research universities and Silicon Valley, is an obvious leader. China, with steely state determination, global linkages and deep resources, is an emerging challenger. Japan, the land of robots, is a leader in automation and related technologies. Russia, which features in the cyber-security nightmares of rival nations, is believed to be investing deeply in AI.

“Whoever leads the AI will lead the world, Vladimir Putin has declared unambiguously. The UK wants to focus on the ethical aspects of AI. UK is home to the influential Oxford philosopher Nick Bostrom, whose work and warnings on super-intelligence has helped bring AI into greater public knowledge.

“AI is a crucial tool to optimize business operations while for countries, it provides both commercial and potentially national security gains like enhanced cyber-security,” said Paul S Triolo, practice director (GeoCyber) at the Eurasia Group.

A gathering momentum on AI in India can now be observed at various levels.

This week, the ministry of external affairs held a closeddoor meeting of global AI experts to discuss how to attract Indian diaspora. Big Indian companies such as Bharti Airtel and Reliance Jio are setting up AI labs. Global corporations NVIDIA, Microsoft and Google have all set up labs in India that are advanced AI-linked work. Indian IT services giants such as Infosys and Wipro are belatedly investing in the space. IT industry body Nasscom is setting up Centre of Excellence (COE) in AI in Karnataka and Telangana with companies like IBM, Microsoft, NVIDIA, Intel and AWS, on the lines of its successful 10,000 Startups program.

There are startups that are joining the AI gold rush, too. SatSure Ltd, launched last year by former Goldman Sachs executive Amardeep Sibia, is one such. It builds AI applications using satellite images to provide real-time data on crop health, yield forecast etc. By some estimates, there are 300-plus startups in India developing or deploying AIrelated technologies.


In academia, Bennett University (backed by The Times of India Group, which publishes The Economic Times) is pitching itself as India’s AI-focused university. It recently organised leadingindia.ai, a conclave for AI-related startups. It’s also an ongoing initiative that aims to skill a million people in AI-related technologies in two years.

“We should not unduly worry about talent shortage. For most problems, 90% of the solution does not require AI expertise, says Kamakoti. State governments like Andhra Pradesh too are joining in.

“We want to build our own Centre for Excellence and groom AI specialists,” says Subbarao Ghanta, IT advisor to Andhra Pradesh chief minister Chandrababu Naidu.

It is in talks with Stanford University and the machine-learning expert Andrew NG to get its strategy in place. In three years, it plans to train 250,000 AI specialists. Extensive efforts are underway to collect data. For example, the state government plans to train 25,000 drone pilots in the next 18 months to collect village- and farm-level data for water level and crop cultivation.

“In AI, there is no insurmountable first mover advantage,” says Prof Kambhampati. In other words, it’s not too late yet for India on the AI playground.