India's AI Regulation Dilemma: Balancing Innovation and Oversight

India Needs a "Whole of Govt" Approach to AI Regulation: Excerpts From a Research Paper - MEDIANAMA

This is an excerpt from a researcher paper titled "India's Advance on AI Regulation" by Amlan Mohanty and Shatakratu Sahu from Carnegie India. To access the full paper, click here. Introduction Artificial intelligence (AI) is a general-purpose technology that has existed since the early 1950s. Its trajectory is marked by cycles of hype and innovation, followed by periods of stagnation and disillusionment. In 2022 alone, more than 30 laws related to AI were passed in over 100 countries. What explains this sudden rush to regulate AI? Some say the launch of ChatGPT in late 2022 was a defining moment. It brought generative AI to the forefront, and along with it, concerns about bias, misinformation, copyright violations, and the impact on labor markets. One might also point to a confluence of factors -- the massive breakthroughs in machine learning, new and powerful capabilities of large language models, and the global reach of social media -- which has stoked the fears of policymakers and prompted new regulations in some countries. And yet, nobody seems to have a clear collective vision for how AI should be regulated. This has resulted in divergent approaches around the world -- from comprehensive legislation in the European Union (EU) to technology-specific rules in China, and voluntary commitments in the United States. Despite these differences, global policymakers seem to agree on one thing -- we must leverage the power of AI while mitigating its risks. Where does this leave India on AI regulation? The existing body of literature concerning India's approach to regulating AI is disconnected, narrow, or superficial. It includes news coverage of regulatory proposals; brief commentaries on national and global policy developments; editorials on what India's approach should be; summaries of the legal landscape; readouts of roundtable discussions; and analyses of specific regulatory issues involving AI. What is missing is a clear and comprehensive analysis of India's overall advance on AI regulation. The goal of this paper is to answer two main questions: The paper is divided into four parts: Part I provides an overview of the current sentiment in government, industry, and civil society in India on the topic of AI regulation. Part II explores the scope and objectives of AI regulation, the nature of AI risks, and areas where additional regulations may be required. Part III examines global approaches to AI regulation and views in India. Part IV suggests a policy roadmap for India on AI regulation. [.................] General Sentiments Across India's government, industry, and civil society, there is broad agreement that: However, there is disagreement about the nature and novelty of AI risks, the extent to which current laws can deal with AI risks, whether or not self-regulation can sufficiently address the risks of AI, and the types of binding rules that are required and when they should be introduced. This raises some important follow-up questions -- What are the risks of AI? Are they novel? What are the gaps in existing laws? What aspect of AI should we regulate? [................] Conclusion As India's policymakers carefully mull the next steps on AI regulation, the brief pause in this continuing advance offers an opportunity to reflect and readjust, lest policymakers get trapped in path dependency and a mindless rush to regulate. This paper suggests a few ways in which such provisions can be introduced. Regulation encompasses more than just laws. It also includes norms, standards, ethical practices, policy frameworks, institutional oversight, and soft laws. Therefore, we suggest a "whole of government" approach in which sectoral agencies, MeitY, and an inter-ministerial body collaborate in a dynamic fashion. An AISI, designed from the ground up keeping India's unique needs in mind, can also supplement state capacity. Finally, and this is important, the process by which AI regulations are framed must be both participative and inclusive. Not only should the data, models, and applications that power India's AI ecosystem be representative of its culture, but so too should the policy frameworks that shape its future trajectory. To that end, it behooves the government to initiate a series of consultations on this topic before continuing its advance on AI regulation. This is an excerpt from a paper titled "India's Advance on AI Regulation" by Amlan Mohanty and Shatakratu Sahu from Carnegie India. To access the full paper, click here.

Is India Rushing to Regulate AI?

"Hoping for a 100% perfect world in any technology, even AI, is a pointless exercise," says Vivek Abraham, senior director at Salesforce. The Indian government's proactive stance on AI regulations has sparked a debate: Is it premature to discuss policies for a technology that is still in its nascent stages within the country? For a population as diverse as India, the journey to understanding and implementing AI cannot just remain limited to the urban area but needs to be extended to the rural areas as well. At the recently concluded Bengaluru Tech Summit, experts from various fields discussed the challenges and opportunities in regulating AI in India. The discussion highlighted the need for a use-case-driven approach rather than a sector-based or generic approach. India is slowly evolving as an AI player, with significant achievements in sectors like healthcare, agriculture, and education. As reported earlier in 2024, Telangana became India's first state to develop its own AI model. This was one of the early adoptions of AI by any Indian state. However, the level of AI integration still remains limited in the country. Understandably, AI adoption in rural areas is slower due to challenges like limited internet access, low digital literacy, and the absence of regionally relevant datasets. Efforts like AI4Bharat and Microsoft's Karya aim to bridge this gap, but large-scale adoption and implementation are still far from reality. Shekar Sivasubramanian, CEO of Wadhwani AI, emphasised the importance of contextualisation for AI deployment in India. "How much of Odia is there on the internet? How much is there from Meghalaya or Tamil Nadu?" he asked, illustrating the need to create AI solutions that reflect India's linguistic and cultural diversity. Sivasubramanian also expressed that policies should be rooted in the realities of rural and semi-urban India, where the needs and workflows are significantly different from those in high-end urban areas. "How does AI coexist when the presupposed assumption is standardisation?" Engaging in policy discussions at this stage may be too early. Formulating regulations is difficult without a comprehensive understanding and widespread implementation of AI across the nation. Subi Chaturvedi, InMobi's chief corporate affairs and public policy officer said, "The test of good regulation is... are you looking at adoption and proliferation, or are you putting the cart before the horse?" She argued that regulation should come post-innovation and adoption rather than presumed progress. Chaturvedi also questioned if it's even possible to create one-size-fits-all regulations for something as fast-changing as AI. When NITI Aayog released the 'National Strategy for Artificial Intelligence' in 2018, AI was still in its birth stage in India. The reports also identified the challenges that come with adopting AI across focus sectors in India. These included the lack of supporting data ecosystems, low levels of AI research and difficulty translating research into physical applications. It also spoke about the shortage of AI expertise and skilling opportunities. Additional challenges were the high cost of resources and low awareness of AI's benefits for businesses, as well as unclear privacy, security, and ethical regulations. An unattractive intellectual property regime also contributed to the hindrances. However, Chaturvedi appreciated the government's efforts to consult with developers, startups, and policymakers as a promising move towards more inclusive and practical solutions. Addressing these challenges quickly through collaborative efforts, with the government leading, could help position India as a leader in AI, especially when considering the rules for India's diverse and underserved communities. AI differs fundamentally from traditional technologies. Vivek Abraham, senior director at Salesforce, highlighted AI's probabilistic nature, noting that identical prompts can produce varying outputs. "Hoping for a 100% perfect world in any technology, even AI, is a pointless exercise," he said. Abraham suggested that instead of chasing perfection, we should focus on frameworks that ensure safety while accepting the truth that some level of failure is inevitable. This is much like how the aviation and automobile industries work. Sunil Abraham, Meta's public policy director, added that AI's unpredictability, including its ability to sometimes produce false answers based on user prompts, makes it tricky to regulate. He recommended tailoring regulations to specific sectors, with stricter rules for high-risk areas, such as medical diagnostics, and leaving creative and low-risk uses unregulated. "Somebody using an LLM on their phone to write a new poem should be unregulated. Similarly, somebody using it to write a script for a movie or the lyrics for a song, should, ideally, be unregulated," he told the audience. The disparity between AI adoption in urban (India) and rural (Bharat) areas is still very large. Urban centres are witnessing AI-driven innovations in fintech and smart cities, whereas rural regions are in need of simple, intuitive, and empathetic solutions. Sivasubramanian highlighted the importance of addressing this divide, urging policymakers to consider the realities of farmers, small businesses, and rural users. He expressed that it is crucial to first create an AI ecosystem in the country, ensuring that both 'India' and 'Bharat' are extensively adopting AI in workflows. "The policy is also applicable to a farmer in Gujarat growing cotton," he said. Policies must be understandable by all those who are relevant, even those who do not understand said technology. "AI knows no jurisdictional borders," said Charmaine Ng, director for Asia Pacific at Schneider Electric. She emphasised the need for international standards that work across borders, highlighting the importance of collaboration among governments, industries, and civil society. Building an ecosystem that supports education, infrastructure, and localised AI solutions is essential. As AI grows and spreads globally, a collated regulation approach that encompasses the needs of all is needed. Ng called for global partnerships to protect privacy, trust, and democracy while helping AI benefit everyone. Such cooperation could also align India's policies with global standards, giving Indian AI innovators a stronger chance to compete internationally.

Is India Choking Innovation by Rushing to Regulate AI?

"Hoping for a 100% perfect world in any technology, even AI, is a pointless exercise," says Vivek Abraham, senior director at Salesforce. AI regulation in India is quickly becoming a hot topic, but are these discussions worth the hype? With AI still in its infancy across the country, many argue that these conversations are out of sync with ground realities. For a nation as diverse as India, true progress in AI must go beyond urban-centric debates and address the unique challenges and opportunities in rural areas. Are we putting the cart before the horse? At the recently concluded Bengaluru Tech Summit, experts from various fields discussed the challenges and opportunities in regulating AI in India. The discussion highlighted the need for a use-case-driven approach rather than a sector-based or generic approach. India is slowly evolving as an AI player, with significant achievements in sectors like healthcare, agriculture, and education. As reported earlier in 2024, Telangana became India's first state to develop its own AI model. This was one of the early adoptions of AI by any Indian state. However, the level of AI integration still remains limited in the country. Understandably, AI adoption in rural areas is slower due to challenges like limited internet access, low digital literacy, and the absence of regionally relevant datasets. Efforts like AI4Bharat and Microsoft's Karya aim to bridge this gap, but large-scale adoption and implementation are still far from reality. Shekar Sivasubramanian, CEO of Wadhwani AI, emphasised the importance of contextualisation for AI deployment in India. "How much of Odia is there on the internet? How much is there from Meghalaya or Tamil Nadu?" he asked, illustrating the need to create AI solutions that reflect India's linguistic and cultural diversity. Sivasubramanian also expressed that policies should be rooted in the realities of rural and semi-urban India, where the needs and workflows are significantly different from those in high-end urban areas. "How does AI coexist when the presupposed assumption is standardisation?" Engaging in policy discussions at this stage may be too early. Formulating regulations is difficult without a comprehensive understanding and widespread implementation of AI across the nation. Subi Chaturvedi, InMobi's chief corporate affairs and public policy officer said, "The test of good regulation is... are you looking at adoption and proliferation, or are you putting the cart before the horse?" She argued that regulation should come post-innovation and adoption rather than presumed progress. Chaturvedi also questioned if it's even possible to create one-size-fits-all regulations for something as fast-changing as AI. When NITI Aayog released the 'National Strategy for Artificial Intelligence' in 2018, AI was still in its birth stage in India. The reports also identified the challenges that come with adopting AI across focus sectors in India. These included the lack of supporting data ecosystems, low levels of AI research and difficulty translating research into physical applications. It also spoke about the shortage of AI expertise and skilling opportunities. Additional challenges were the high cost of resources and low awareness of AI's benefits for businesses, as well as unclear privacy, security, and ethical regulations. An unattractive intellectual property regime also contributed to the hindrances. However, Chaturvedi appreciated the government's efforts to consult with developers, startups, and policymakers as a promising move towards more inclusive and practical solutions. Addressing these challenges quickly through collaborative efforts, with the government leading, could help position India as a leader in AI, especially when considering the rules for India's diverse and underserved communities. AI differs fundamentally from traditional technologies. Vivek Abraham, senior director at Salesforce, highlighted AI's probabilistic nature, noting that identical prompts can produce varying outputs. "Hoping for a 100% perfect world in any technology, even AI, is a pointless exercise," he said. Abraham suggested that instead of chasing perfection, we should focus on frameworks that ensure safety while accepting the truth that some level of failure is inevitable. This is much like how the aviation and automobile industries work. Sunil Abraham, Meta's public policy director, added that AI's unpredictability, including its ability to sometimes produce false answers based on user prompts, makes it tricky to regulate. He recommended tailoring regulations to specific sectors, with stricter rules for high-risk areas, such as medical diagnostics, and leaving creative and low-risk uses unregulated. "Somebody using an LLM on their phone to write a new poem should be unregulated. Similarly, somebody using it to write a script for a movie or the lyrics for a song, should, ideally, be unregulated," he told the audience. The disparity between AI adoption in urban (India) and rural (Bharat) areas is still very large. Urban centres are witnessing AI-driven innovations in fintech and smart cities, whereas rural regions are in need of simple, intuitive, and empathetic solutions. Sivasubramanian highlighted the importance of addressing this divide, urging policymakers to consider the realities of farmers, small businesses, and rural users. He expressed that it is crucial to first create an AI ecosystem in the country, ensuring that both 'India' and 'Bharat' are extensively adopting AI in workflows. "The policy is also applicable to a farmer in Gujarat growing cotton," he said. Policies must be understandable by all those who are relevant, even those who do not understand said technology. "AI knows no jurisdictional borders," said Charmaine Ng, director for Asia Pacific at Schneider Electric. She emphasised the need for international standards that work across borders, highlighting the importance of collaboration among governments, industries, and civil society. Building an ecosystem that supports education, infrastructure, and localised AI solutions is essential. As AI grows and spreads globally, a collated regulation approach that encompasses the needs of all is needed. Ng called for global partnerships to protect privacy, trust, and democracy while helping AI benefit everyone. Such cooperation could also align India's policies with global standards, giving Indian AI innovators a stronger chance to compete internationally.

AI Regulation Talks in India Are Useless Right Now

"Hoping for a 100% perfect world in any technology, even AI, is a pointless exercise," says Vivek Abraham, senior director at Salesforce. AI regulation in India is quickly becoming a hot topic, but are these discussions worth the hype? With AI still in its infancy across the country, many argue that these conversations are out of sync with ground realities. For a nation as diverse as India, true progress in AI must go beyond urban-centric debates and address the unique challenges and opportunities in rural areas. Are we putting the cart before the horse? At the recently concluded Bengaluru Tech Summit, experts from various fields discussed the challenges and opportunities in regulating AI in India. The discussion highlighted the need for a use-case-driven approach rather than a sector-based or generic approach. India is slowly evolving as an AI player, with significant achievements in sectors like healthcare, agriculture, and education. As reported earlier in 2024, Telangana became India's first state to develop its own AI model. This was one of the early adoptions of AI by any Indian state. However, the level of AI integration still remains limited in the country. Understandably, AI adoption in rural areas is slower due to challenges like limited internet access, low digital literacy, and the absence of regionally relevant datasets. Efforts like AI4Bharat and Microsoft's Karya aim to bridge this gap, but large-scale adoption and implementation are still far from reality. Shekar Sivasubramanian, CEO of Wadhwani AI, emphasised the importance of contextualisation for AI deployment in India. "How much of Odia is there on the internet? How much is there from Meghalaya or Tamil Nadu?" he asked, illustrating the need to create AI solutions that reflect India's linguistic and cultural diversity. Sivasubramanian also expressed that policies should be rooted in the realities of rural and semi-urban India, where the needs and workflows are significantly different from those in high-end urban areas. "How does AI coexist when the presupposed assumption is standardisation?" Engaging in policy discussions at this stage may be too early. Formulating regulations is difficult without a comprehensive understanding and widespread implementation of AI across the nation. Subi Chaturvedi, InMobi's chief corporate affairs and public policy officer said, "The test of good regulation is... are you looking at adoption and proliferation, or are you putting the cart before the horse?" She argued that regulation should come post-innovation and adoption rather than presumed progress. Chaturvedi also questioned if it's even possible to create one-size-fits-all regulations for something as fast-changing as AI. When NITI Aayog released the 'National Strategy for Artificial Intelligence' in 2018, AI was still in its birth stage in India. The reports also identified the challenges that come with adopting AI across focus sectors in India. These included the lack of supporting data ecosystems, low levels of AI research and difficulty translating research into physical applications. It also spoke about the shortage of AI expertise and skilling opportunities. Additional challenges were the high cost of resources and low awareness of AI's benefits for businesses, as well as unclear privacy, security, and ethical regulations. An unattractive intellectual property regime also contributed to the hindrances. However, Chaturvedi appreciated the government's efforts to consult with developers, startups, and policymakers as a promising move towards more inclusive and practical solutions. Addressing these challenges quickly through collaborative efforts, with the government leading, could help position India as a leader in AI, especially when considering the rules for India's diverse and underserved communities. AI differs fundamentally from traditional technologies. Vivek Abraham, senior director at Salesforce, highlighted AI's probabilistic nature, noting that identical prompts can produce varying outputs. "Hoping for a 100% perfect world in any technology, even AI, is a pointless exercise," he said. Abraham suggested that instead of chasing perfection, we should focus on frameworks that ensure safety while accepting the truth that some level of failure is inevitable. This is much like how the aviation and automobile industries work. Sunil Abraham, Meta's public policy director, added that AI's unpredictability, including its ability to sometimes produce false answers based on user prompts, makes it tricky to regulate. He recommended tailoring regulations to specific sectors, with stricter rules for high-risk areas, such as medical diagnostics, and leaving creative and low-risk uses unregulated. "Somebody using an LLM on their phone to write a new poem should be unregulated. Similarly, somebody using it to write a script for a movie or the lyrics for a song, should, ideally, be unregulated," he told the audience. The disparity between AI adoption in urban (India) and rural (Bharat) areas is still very large. Urban centres are witnessing AI-driven innovations in fintech and smart cities, whereas rural regions are in need of simple, intuitive, and empathetic solutions. Sivasubramanian highlighted the importance of addressing this divide, urging policymakers to consider the realities of farmers, small businesses, and rural users. He expressed that it is crucial to first create an AI ecosystem in the country, ensuring that both 'India' and 'Bharat' are extensively adopting AI in workflows. "The policy is also applicable to a farmer in Gujarat growing cotton," he said. Policies must be understandable by all those who are relevant, even those who do not understand said technology. "AI knows no jurisdictional borders," said Charmaine Ng, director for Asia Pacific at Schneider Electric. She emphasised the need for international standards that work across borders, highlighting the importance of collaboration among governments, industries, and civil society. Building an ecosystem that supports education, infrastructure, and localised AI solutions is essential. As AI grows and spreads globally, a collated regulation approach that encompasses the needs of all is needed. Ng called for global partnerships to protect privacy, trust, and democracy while helping AI benefit everyone. Such cooperation could also align India's policies with global standards, giving Indian AI innovators a stronger chance to compete internationally.