Niti Aayog’s National AI strategy to have two tiered structure for research and prototype development
It proposes the creation of Centers of Research Excellence (CoREs) for fundamental research, and International Centers for Transformational AI (ICTAI) for applied research. EC Speaks with Anna Roy, Advisor, Industry and Data Management & Analysis verticals, NITI Aayog
The NITI Aayog’s CEO Amitabh Kant had written to all states and ministries requesting them to identify key projects across the five identified sectors where AI can be adopted. “To give a push to adoption of AI, it is pertinent to have an AI check in all new projects to identify specific areas where AI can be used,” Kant said in a letter dated May 7, 2019.
Which are the major social issues that the states want to address using AI based on the responses received from them?
At this stage, state governments are interested in use cases across healthcare, agriculture, public safety, smart city and environment. In response to the letters, state governments have mainly asked for support in:
○ Identification of potential AI use cases
○ Connection to experts in various sub-domains of AI application
○ Means of preparing data for AI usage
Please enumerate the ministries and departments (whichever applies), where Artificial Intelligence (AI) has been / in the process of or is being explored to be applied?
Large scale adoption of AI within the government is limited, though we are working to help central and state governments identify potential use cases and increase adoption. Some of the major applications of AI being explored/in implementation in NITI Aayog are as follows:
NITI Aayog is exploring the usage of AI for ‘assaying’ of quality of fruits and vegetables in eNAM integrated mandis in coordination with the Government of Telangana;
We are working with the Department of Biotechnology towards the creation of a ‘biobank’ of pathological and radiological cancer images that may be used for the development of sophisticated AI solutions;
NITI Aayog is developing the National Data and Analytics Platform. AI tools will be leveraged to deliver a platform that will democratize access to public government data through a user-friendly interface
Healthcare, education, urban development, agriculture are the areas where the Government wants to pursue AI on a mission mode? Please share more.
AI has potential for adoption across all sectors. However, for a more focused Government intervention and leveraging limited public resources a more nuanced approach for government intervention has been recommended in the National Strategy on Artificial Intelligence. The strategy advocates that sectors such as healthcare, education, agriculture, urban development, and mobility may be the focus sector for government support for adoption of AI since it may not be economically attractive for private sector entities to invest in research and solutions alone.
The National Strategy recommends that a two tiered structure for research and prototype development be created to ensure the translation of fundamental research in AI into commercially viable products. It proposes the creation of Centers of Research Excellence (CoREs) for fundamental research, and International Centers for Transformational AI (ICTAI) for applied research. It is envisaged that ICTAIs may be sector specific, and responsible for creation of products for ‘social good’ use cases in these sectors.
It’s imperative to build an ecosystem for AI to make it scalable to be applied across use cases? What steps have the Government taken for the same?
NITI Aayog understands that the government will be a major adopter of AI technology. However, the primary challenge is the identification of viable use cases and the gathering of data required for implementation. NITI Aayog has conducted 2 workshops with State Governments to help them identify potential applications of AI, and the underlying infrastructure required for their implementation. The themes of the workshop included ‘the need for structured data’, ‘designing moonshot challenges for AI’, etc. India also has a growing AI ecosystem as startups and private companies expand their AI capabilities and develop new tools.
Globally, what is your assessment on the standing that India holds in AI preparedness?
The National Strategy on AI envisages to make India a leader in the space of ‘AI for All’ or ‘AI for Social Good’. As India’s internet and smartphone penetration continues to grow – it is envisaged that India will transition to a major market for AI driven products and solutions. It is believed that as this transition occurs India is well placed to become a leader in the usage of AI in social sectors – given it’s large diversity and varying demographic.
NSAI recommends positioning India as an AI garage of the world. India presents use cases in various sectors aimed at addressing problems faced by 40% of global population. Solving for India can help solving for the world. India can be the test bed for developing and stabilizing AI tools that can then have a global market.
India lacks in the number of AI specialists. What is your take?
It has been observed that there is a lack of AI researchers and AI specialists in India. However there are some positive factors that indicate that this trend is reversing:
Many of the renowned AI practitioners in western countries are of Indian origin and are very keen to contribute back to society in India through allocation of resources and time for research and solution development. Leveraging the capacity of the Indian diaspora would benefit not just solution development, but also the transfer of technology and know-how to create our own pool of specialists in the near future.
Many of the students who are currently studying in undergraduate courses or students who have graduated recently are very keen to specialize in the field of AI (or machine learning) due to its perceived demand. Though they are currently in early parts of their careers, it is believed that they will rise to prominence as specialists soon
Additionally, there is a significant increase in AI courses which are delivered through the web as MooCs or other mediums. This is allowing an opportunity for professionals to expand their skillset to cover AI related specialization.
Which are the specific areas where more AI startups need to come up?
NLP for Indic languages – especially in speech to text and text to speech areas. This is one area in which it is essential that startups take the lead, due to the degree of specialized domestic knowledge required for execution.
Solutions relating to the areas of healthcare, agriculture, public safety, and environment can create tremendous social impact and will help India become a leader in the space of ‘AI for All’ or ‘AI for Social Good’
Use of AI in security. Where does India stand?
India is currently exploring the usage of AI in National Security. A Task Force is constituted specifically to explore potential applications of AI in Defense. The report of the Task Force has been submitted to the Ministry of Defense.
