Industry experts are urging New Delhi to design a home-grown artificial intelligence safety framework suited to India's languages, scale and application-led AI ecosystem, rather than replicating the voluntary pact US President Donald Trump signed with top technology firms last week.
The comments follow the White House Accord on Super Intelligence, signed on September 29 by Trump and executives from Google, OpenAI, Anthropic, Meta, Nvidia and Elon Musk's xAI. Under the accord, the companies committed to internal controls for monitoring their models, dedicated oversight teams, independent external audits, board-level oversight and regular meetings to set safety standards.
US model "not a template" for India
Nasscom Chairman and Fractal Group CEO Srikanth Velamakanni said India should not simply copy the US approach, and could instead use a domestic commitment as a building block toward a global understanding of AI safety.
"As models become more capable and agentic, we are giving them greater access to systems and greater autonomy to make decisions. AI safety must be strengthened considerably and must stay ahead of capability to prevent catastrophic risks," he said.
Velamakanni noted that the US accord is largely a push for self-policing by frontier AI companies, while the European Union has taken a binding, risk-based approach and China follows a state-directed control model. India, he said, could use its role in the Global Partnership on AI and the momentum from hosting the AI Impact Summit in February to help prevent worst-case outcomes without the Global South being denied access to frontier models.
Split on self-regulation
Soumen Mandal, Principal Analyst at Counterpoint Research, said India could take inspiration from the US accord and move faster to build a trusted AI ecosystem. He said such a commitment should cover testing AI systems before deployment, independent audits, reporting of serious problems and naming the people responsible for AI-related risks.
Amit Khanna, Partner and Automation Ecosystem Leader at Grant Thornton Bharat, said any Indian commitment must suit local conditions. India's focus, he noted, is on building AI applications rather than major foundational models, and the rules must ensure AI works reliably across vernacular languages and cultural contexts without exploiting the population's AI knowledge gap. Gnani.ai co-founder and CEO Ganesh Gopalan added that global frontier models should face the same rules as Indian ones, since Indian AI companies compete with global firms for the same customers.
The experts differed on whether self-regulation would be enough. Velamakanni called voluntary approaches "necessary but woefully insufficient," pointing to pharmaceuticals and aviation as industries where strong, technically competent regulators strengthened safety and trust without stopping innovation. He said India needs binding rules on transparency, incident reporting and accountability when serious harm occurs, and called for strengthening the IndiaAI Safety Institute to test newly released models — without requiring pre-release certification of every model, which would create a bottleneck.
Mandal backed a hybrid approach: mandatory rules for potential harms in defence, healthcare, banking and government use cases, with voluntary standards for fast-moving technical areas where regulators may lack expertise. Khanna also called for a hybrid "techno-legal" approach, pairing a strong voluntary layer with binding mandates such as the provisions on synthetically generated information in the IT Amendment Rules, 2026. Gopalan, however, favoured self-regulation, warning that mandatory regulation could stifle competition and saying India needs more than 100 frontier AI lab companies to avoid falling behind in the AI race.
The experts named deepfakes as an immediate priority. Khanna described misinformation, deepfakes and political manipulation as "existential threats to India's social fabric," while Velamakanni listed large-scale cyberattacks, electoral interference, biological or chemical weapons, attacks on critical infrastructure and loss of control over highly autonomous systems as the top concerns. "Autonomy must lag behind assurance to provide a margin of safety," he said.
The Indian government is already moving in this direction. IT Secretary S Krishnan said earlier this year that while existing legal provisions had been adequate for initial concerns such as deepfakes and AI-generated synthetic content, an "additional regulation or law may be needed." IT Minister Ashwini Vaishnaw has similarly said the current IT law was framed before AI's rapid emergence and a new legal framework may be required.
Sources: India Needs Its Own AI Safety Framework, Not A Copy Of US Pact: Industry Experts — ETV Bharat/PTI
The comments follow the White House Accord on Super Intelligence, signed on September 29 by Trump and executives from Google, OpenAI, Anthropic, Meta, Nvidia and Elon Musk's xAI. Under the accord, the companies committed to internal controls for monitoring their models, dedicated oversight teams, independent external audits, board-level oversight and regular meetings to set safety standards.
US model "not a template" for India
Nasscom Chairman and Fractal Group CEO Srikanth Velamakanni said India should not simply copy the US approach, and could instead use a domestic commitment as a building block toward a global understanding of AI safety.
"As models become more capable and agentic, we are giving them greater access to systems and greater autonomy to make decisions. AI safety must be strengthened considerably and must stay ahead of capability to prevent catastrophic risks," he said.
Velamakanni noted that the US accord is largely a push for self-policing by frontier AI companies, while the European Union has taken a binding, risk-based approach and China follows a state-directed control model. India, he said, could use its role in the Global Partnership on AI and the momentum from hosting the AI Impact Summit in February to help prevent worst-case outcomes without the Global South being denied access to frontier models.
Split on self-regulation
Soumen Mandal, Principal Analyst at Counterpoint Research, said India could take inspiration from the US accord and move faster to build a trusted AI ecosystem. He said such a commitment should cover testing AI systems before deployment, independent audits, reporting of serious problems and naming the people responsible for AI-related risks.
Amit Khanna, Partner and Automation Ecosystem Leader at Grant Thornton Bharat, said any Indian commitment must suit local conditions. India's focus, he noted, is on building AI applications rather than major foundational models, and the rules must ensure AI works reliably across vernacular languages and cultural contexts without exploiting the population's AI knowledge gap. Gnani.ai co-founder and CEO Ganesh Gopalan added that global frontier models should face the same rules as Indian ones, since Indian AI companies compete with global firms for the same customers.
The experts differed on whether self-regulation would be enough. Velamakanni called voluntary approaches "necessary but woefully insufficient," pointing to pharmaceuticals and aviation as industries where strong, technically competent regulators strengthened safety and trust without stopping innovation. He said India needs binding rules on transparency, incident reporting and accountability when serious harm occurs, and called for strengthening the IndiaAI Safety Institute to test newly released models — without requiring pre-release certification of every model, which would create a bottleneck.
Mandal backed a hybrid approach: mandatory rules for potential harms in defence, healthcare, banking and government use cases, with voluntary standards for fast-moving technical areas where regulators may lack expertise. Khanna also called for a hybrid "techno-legal" approach, pairing a strong voluntary layer with binding mandates such as the provisions on synthetically generated information in the IT Amendment Rules, 2026. Gopalan, however, favoured self-regulation, warning that mandatory regulation could stifle competition and saying India needs more than 100 frontier AI lab companies to avoid falling behind in the AI race.
The experts named deepfakes as an immediate priority. Khanna described misinformation, deepfakes and political manipulation as "existential threats to India's social fabric," while Velamakanni listed large-scale cyberattacks, electoral interference, biological or chemical weapons, attacks on critical infrastructure and loss of control over highly autonomous systems as the top concerns. "Autonomy must lag behind assurance to provide a margin of safety," he said.
The Indian government is already moving in this direction. IT Secretary S Krishnan said earlier this year that while existing legal provisions had been adequate for initial concerns such as deepfakes and AI-generated synthetic content, an "additional regulation or law may be needed." IT Minister Ashwini Vaishnaw has similarly said the current IT law was framed before AI's rapid emergence and a new legal framework may be required.
Sources: India Needs Its Own AI Safety Framework, Not A Copy Of US Pact: Industry Experts — ETV Bharat/PTI
