India Urged to Build Its Own AI Safety Framework, Not Copy US Pact

India’s technology leaders are pushing for a bespoke artificial intelligence (AI) safety framework, arguing that the country’s linguistic diversity, large-scale applications, and democratic context require tailored solutions. The debate follows the White House Accord on Super Intelligence, a voluntary pact signed by major US tech firms, and highlights growing concerns about risks like deepfakes, cyberattacks, and regulatory gaps.

India’s Push for a Homegrown AI Safety Framework

Industry leaders in India are urging the country to develop its own artificial intelligence (AI) safety framework, arguing that a one-size-fits-all approach modeled after the US or European Union may not suit its needs. According to YourStory, experts emphasize that India’s AI ecosystem is shaped by its linguistic diversity, large-scale applications, and a focus on deploying AI rather than building foundational models. This context, they say, demands a bespoke regulatory approach.

The discussion follows the White House Accord on Super Intelligence, a voluntary pact signed by major US technology firms, including Google, OpenAI, Anthropic, Meta, Nvidia, and xAI. The accord relies on self-policing, with companies committing to internal controls, independent audits, and board-level oversight. However, Indian experts caution that this model may not be sufficient for India’s unique challenges.

Debating Voluntary vs. Binding Regulation

There is growing debate among Indian experts about whether voluntary self-regulation or binding rules would better address AI-related risks. Srikanth Velamakanni, Chairman of Nasscom and CEO of Fractal Group, argued that India should not simply replicate the US framework. Instead, he suggested drawing inspiration from it while tailoring solutions to India’s specific needs.

Amit Khanna, Partner and Automation Ecosystem Leader at Grant Thornton Bharat, emphasized that any commitment to AI safety must align with Indian market conditions. He described misinformation, deepfakes, and political manipulation as "existential threats to India’s social fabric," underscoring the need for robust safeguards.

Soumen Mandal, Principal Analyst at Counterpoint Research, proposed a hybrid approach. He suggested combining mandatory regulations for high-risk sectors like defense, healthcare, and banking with voluntary standards for rapidly evolving technical areas. This, he argued, would balance safety with innovation.

Key Risks and Regulatory Gaps

Experts highlighted several risks that India’s AI safety framework must address. Deepfakes, cyberattacks targeting financial institutions, and large-scale misinformation were identified as top concerns, particularly given India’s democratic framework and large population. Ganesh Gopalan, Co-founder and CEO of Gnani.ai, noted that these risks require urgent attention to prevent systemic harm.

Another critical gap is accountability. Rules on liability—such as when an AI system misdiagnoses a patient or wrongly denies welfare benefits—remain undeveloped. Mandal prioritized independent audits, followed by accountability, data protection, and job impact as key focus areas for AI safety.

The lack of clear liability frameworks was flagged as a significant vulnerability. Without binding rules on transparency, incident reporting, and accountability, experts warn that India could face repeat incidents of AI-driven harm with no recourse for affected individuals or businesses.

Government’s Role and Next Steps

The Indian government is actively considering new AI regulations. IT Minister Ashwini Vaishnaw and IT Secretary S Krishnan have both signaled the need for an updated legal framework. Vaishnaw noted that the existing information technology law predates the rapid emergence of AI, while Krishnan suggested that additional regulations may be necessary to address evolving risks like deepfakes and synthetic content.

A multi-stakeholder approach is widely seen as essential for effective AI governance. Experts argue that any framework should include government agencies, frontier AI developers, enterprises, academia, civil society, and independent researchers. Soumen Mandal reiterated this, stating that an AI safety agreement must bring together the entire ecosystem—from AI companies to end-users.

India’s participation in the Global Partnership on AI (GPAI) and its recent hosting of the AI Impact Summit were cited as opportunities to shape global AI governance. However, experts stress that India must prioritize its own requirements while engaging in these forums.

Context: Global AI Regulation Trends

India’s deliberations on AI safety come as governments worldwide grapple with regulating the technology. The European Union has adopted a binding, risk-based approach, while China relies on state-directed control. The US, in contrast, leans toward voluntary commitments from tech companies, as seen in the White House Accord on Super Intelligence.

Recent incidents have highlighted the urgency of AI regulation. Security researchers uncovered vulnerabilities in AI agents, such as prompt injection attacks via email, while regulatory actions—like India’s removal of the app Bitchat from its App Store—demonstrate growing scrutiny of AI-driven applications.

The debate in India mirrors global discussions about balancing innovation with safety. While self-regulation offers flexibility, critics argue it leaves gaps in civil rights and public safety, particularly in high-stakes sectors like healthcare and finance.

What this means

Lazyfounder analysis — our interpretation, not reported fact.

India’s push for a homegrown AI safety framework reflects a broader tension in global tech policy: how to regulate cutting-edge technology without stifling innovation or sovereignty. The debate pits pragmatists, who argue for tailored rules that address India’s unique challenges—like its linguistic diversity and application-driven AI ecosystem—against those wary of over-regulation slowing down its tech growth.

The US model of self-regulation, while flexible, may not address the risks of AI-driven misinformation, cyber threats, or systemic failures that could destabilize India’s democracy or economy. A binding, multi-stakeholder approach could offer stronger safeguards, but it risks creating bottlenecks if not designed carefully. For founders and operators, this uncertainty underscores the need to proactively engage in policy discussions, invest in compliance infrastructure, and stay agile as regulations evolve.

The real test will be whether India can strike a balance—protecting citizens and businesses from AI’s risks while ensuring its startups and enterprises remain competitive globally. The conversation is no longer theoretical; it’s a signal to prepare for regulatory change.

Key takeaways

• India’s AI safety framework must reflect its unique linguistic, cultural, and technological context rather than mirroring Western models.
• A hybrid regulatory approach—combining mandatory rules for high-risk sectors with voluntary standards for rapid innovation—could balance safety and growth.
• Deepfakes, cyberattacks, and misinformation are critical concerns, requiring clear liability rules and independent audits for AI systems.
• Government, industry, academia, and civil society must collaborate to design effective AI governance, leveraging forums like the Global Partnership on AI (GPAI).
• Over-regulation risks stifling competition and innovation, while under-regulation could leave gaps in civil rights and public safety protections.

FAQ

Why does India need its own AI safety framework?

Experts argue that India’s AI ecosystem is unique due to its linguistic diversity, large-scale applications, and focus on deploying AI rather than building foundational models. A tailored framework would address these specific challenges and risks, such as deepfakes and cyberattacks, more effectively than a one-size-fits-all approach.

What is the White House Accord on Super Intelligence?

The White House Accord on Super Intelligence is a voluntary pact signed by major US technology firms, including Google, OpenAI, Anthropic, Meta, Nvidia, and xAI. It relies on self-regulation, with companies committing to internal controls, independent audits, and board-level oversight of AI safety.

What are the main risks India’s AI safety framework should address?

Key risks include deepfakes, cyberattacks targeting financial institutions, misinformation, and systemic failures like large-scale cyberattacks or loss of control over autonomous AI systems. Experts also highlight the need for clear liability rules when AI systems cause harm.

What is the debate between voluntary and binding AI regulations?

Voluntary regulations, like those in the US, offer flexibility but may leave gaps in civil rights and public safety. Binding rules, such as those in the European Union, provide stronger safeguards but could stifle innovation if overly restrictive. Experts in India are debating a hybrid approach to balance these concerns.

What role is the Indian government playing in AI regulation?

The Indian government is actively considering new AI regulations. IT Minister Ashwini Vaishnaw and IT Secretary S Krishnan have both indicated that the existing legal framework may need updating to address risks like deepfakes and synthetic content. A multi-stakeholder approach is expected to shape the final regulations.

Related on Lazyfounder

• More ai news
• OpenAI Safety Researcher Resigns, Warns of Reckless AI Development