Centre's Rs 10,300 Cr Allocation For AI Mission: Experts Weigh In

The Cabinet has approved an allocation of over Rs 10,300 crore for the IndiaAI Mission in an effort to support India’s artificial intelligence (AI) ecosystem. The financial infusion is planned over the next five years and is aimed at accelerating various components of the IndiaAI Mission, with a strong emphasis on public-private partnerships to encourage innovation and entrepreneurship in the field of AI.

The IndiaAI mission will be spearheaded by the Independent Business Division (IBD) under Digital India Corporation (DIC), which will look to establish a comprehensive ecosystem that drives AI innovation through strategic programs and partnerships across the public and private sectors. 

The mission encompasses several key components aimed at democratising computing access, improving data quality, developing indigenous AI capabilities, attracting top AI talent, enabling industry collaboration, providing startup risk capital, ensuring socially impactful AI projects and bolstering ethical AI practices.

“AI is going to be the kinetic enabler for India’s digital economy. Prime Minister Modi ji has always said that India is going to fully exploit the power of AI for the benefit of its citizens and for the expansion of its economy. I am grateful and thankful to the Hon’ble PM for the approval of more than Rs 10,000 crore for the India AI program, which will catalyse India’s AI ecosystem and position it as a force shaping the future of AI for India and for the world.” - Rajeev Chandrasekhar, Union Minister of State for Electronics and IT

10,000 GPUs Through Public-Private Collaborations

One of the core initiatives under the IndiaAI Mission is the IndiaAI Compute Capacity, which aims to build a high-end scalable AI computing ecosystem to meet the increasing demands from India’s rapidly expanding AI startups and research community. This initiative will involve acquiring (in 18-24 months) and deploying over 10,000 Graphics Processing Units (GPUs) through strategic public-private collaborations, along with the creation of an AI marketplace offering AI as a service and pre-trained models to innovators.

However, the specific models of GPUs which would be acquired for the IndiaAI mission were not explicitly disclosed. It remains uncertain whether the GPUs include Nvidia’s premium data centre offerings, such as the A100 series, priced around Rs 12 lakh each, or the H100 series, which come at approximately Rs 50 lakh each, with costs varying based on packaging configurations. Basis the quality of chips acquired, a substantial chunk of the allocated fund could be solely for this acquisition.

“The proposal to invest into GPU-based compute capacity is a welcome effort to enable the ‘GenAI economy’ for creating Gen AI based applications. This move gives fillip to the startup ecosystem to drive innovation helping India leapfrog into areas such as text summarisation, video/image generation and next gen chatbots etc, including enablement across Indian languages,” said Prashanth Kaddi, Partner, Consulting, Deloitte India.

“It also would help India play the long game on AI given the public sector participation, as it would help in ensuring ‘productionalisation’ and democratisation of Gen AI applications in the most cost-effective way for adoption at scale (potentially a billion users for citizen services),” he added,

According to the plan laid out by the central government, 'IndiaAI Innovation Centre' will play a pivotal role in developing and deploying indigenous Large Multimodal Models (LMMs) and domain-specific foundational models in critical sectors to enhancing the country’s capabilities in AI research and development. The Innovation Centre is envisioned to be a leading academic institution to focus on streamlined implementation and retention of top research talent.

‘More Can Be Done’

The AI announcement, seemingly absent during the Interim Budget, carries substantial weight with a price tag of Rs 10,300 crore and the provision of 10,000+ GPUs. However, it may merely represent a starting point, considering the ecosystem that it will have to sustain under this endeavour.

Speaking with BW Businessworld, Professor Meir Feder, School of Electrical Engineering, Tel Aviv University and Head, TAD, said, “It (10,000 GPUs) is a nice number. It seems reasonable for academia and research and startups, if clusters are built for them. However, it is not enough if you want to compete with the likes of OpenAI and Anthropic.”

“The Rs 10,300 crore (USD 1.2 billion) outlay is substantial indeed. But I hope this will be directed mainly towards research and helping startups, as it may not be enough for a larger endeavour. If you compare with other larger countries such as US or China, it may not as big a scale.” -  Professor Meir Feder, School of Electrical Engineering, Tel Aviv University and Head, TAD

From his experience, working under the national AI programme in Israel at Tel Aviv University, Feder said that the Indian government should also consider handing out grants among academia and universities to encourage continued innovation. He also suggested that the government should be proactive in checking how the resources are being distributed to prevent bureaucracy and politics from halting access to resources. 

In addition to these recommendations, Feder stressed the importance of India leveraging and bringing in talent from beyond its borders, including individuals of Indian origin who have the necessary experience in the space and also those from diverse backgrounds, to maximise the innovation potential.

Israel’s current AI programme, announced in 2022, is budgeted at one billion shekels, approximately USD 250 million, allocated in two batches of half a billion shekels each. This programme is mostly focused on AI research. “We anticipate a third batch within the next year or year and a half. This batch-oriented approach allows us to adapt to the dynamic nature of the field, focusing our efforts efficiently. The programme commenced in early 2022 and is expected to run until 2026 or 2027,” Ziv Katzir, Head of the National Program for AI Infrastructure at Israel Innovation Authority told BW Businessworld in February. 

Meanwhile, HaiVE Co-founder Arjun Reddy felt that the Indian government’s outlay was too science-oriented. “While the fact that our government is this aggressive on AI is something to appreciate, it does show that the advisors to this mission are only scientific and not AI business-oriented. If they were then they would’ve brought up that the most important part of the AI revolution is using it and for that, they need inference-specific infrastructures,” he said, asking for business-oriented initiatives as well.

Need For Clarity 

Another crucial aspect of the IndiaAI Mission is the ‘IndiaAI Datasets Platform’, aimed at streamlining access to quality non-personal datasets for AI innovation. The platform will develop a unified data platform, providing seamless access to non-personal datasets for Indian startups and researchers, thereby facilitating AI development and research.

Shreya Suri, Partner at INDUSLAW, noted the significant commitment of resources in the IndiaAI mission, but she also highlighted the intriguing mention of an AI datasets platform for providing non-personal data to startups and researchers, while expressing concerns about data collation and its alignment with the Digital India Act proposal. Suri stressed the need for clarity on the scope and management of non-personal data, urging its distinction from personal data under the DPDP Act. 

“While the core pillars remain employment, innovation, responsible AI and AI for development, there is also an interesting mention of an AI data sets platform through which seamless access to non-personal data sets will be made available to Indian startups and researchers. There has been talk of a non-personal data governance framework for some time now. However, it is unclear how this data is expected to be collated and whether this will tie into the government’s plans for the Digital India Act proposal as well. What will also require some thought is the scope of what non-personal data will entail and how it must be kept distinct and separate from the widely defined concept of personal data under the DPDP Act.” - Shreya Suri, Partner at INDUSLAW

Apps And Jobs

Under the AI mission, the Centre said there will be a IndiaAI Application Development Initiative, which will focus on promoting AI applications in critical sectors by addressing problem statements sourced from central ministries, state departments and other institutions. This initiative will look to develop, scale and promote the adoption of impactful AI solutions with the potential for accelerating large-scale socio-economic transformation.

As per findings from an EY report, around 69 per cent of the projected impact from Generative AI is anticipated to stem from various sectors, notably business services encompassing IT, legal, consulting, outsourcing, rental services, among others. Additionally, financial services, transportation and logistics, education, retail trade and healthcare sectors are expected to contribute significantly to this impact.

“India could experience a substantial boost in its GDP over seven years (2023-24 to 2029-30). The cumulative impact on GDP may range from USD 1.2 trillion to USD 1.5 trillion, contributing an additional 0.9 per cent to 1.1 per cent in annual CAGR.” – EY’s "The AIdea of India report"

Lastly, the 'IndiaAI FutureSkills' programme will focus on expanding the reach of AI education by increasing the accessibility of graduate and post-graduate AI programmes. 

During a press conference, Union Minister of Commerce Piyush Goyal on Thursday revealed that under the Skill Development ‘Kaushal Vikas Yojana’, youth will undergo training in AI, with programmes spanning undergraduate, postgraduate and PhD levels. Foundational AI courses will be extended to tier-2 and tier-3 cities as well as remote areas. The mission will extend early-stage and growth-stage financing to emerging deeptech AI startups. Projects will also be initiated to develop tools and frameworks for ensuring the safety and reliability of AI applications.

According to the monthly Naukri JobSpeak Index, there was a noted increase in recruitment for specialised AI positions like Machine Learning (ML) engineers and full-stack AI scientists in India, with a growth of 100 per cent and 44 per cent, respectively, in February year-over-year (YoY). Conversely, while traditional data scientist roles experienced a moderate 15 per cent uptick in hiring activity, the overall IT sector saw a slowdown of 15 per cent in recruitment during the same period.