India working on balanced AI regulatory framework with safeguards, innovation in focus: IT Secretary
New Delhi, July 19
India is developing a carefully calibrated regulatory framework for artificial intelligence that seeks to balance innovation with necessary safeguards, while drawing lessons from evolving approaches adopted by the US, the European Union and other international forums, Electronics and Information Technology Secretary S Krishnan has said.
As AI capabilities continue to advance rapidly across the world, governments are increasingly examining ways to regulate powerful AI models without hindering technological progress.
India is simultaneously expanding its AI ecosystem through the IndiaAI Mission, strengthening public computing infrastructure and encouraging indigenous AI development.
Speaking to NDTV Profit, Krishnan said there is now growing global consensus that AI cannot remain completely unregulated.
"Globally, there's a belief that AI needs to be regulated. The loudest voice against AI regulation was from the US, but today they're regulating access to certain frontier models. That itself shows how the thinking has evolved," he said.
Krishnan said India is closely studying how different jurisdictions are approaching AI governance while ensuring that any domestic regulatory framework remains proportionate, practical and future-ready.
"Our response has to be a well-calibrated set of rules. We need safeguards and guardrails that AI platforms must follow, but at the same time regulation shouldn't stifle innovation," he said.
He acknowledged that designing AI legislation presents unique challenges because of the speed at which the technology is evolving.
"The law has to keep pace with technology. That's a fine line we need to tread and designing such a framework is a challenge. It will take some time," he noted.
The Electronics and IT Secretary said India has been actively participating in international discussions on AI governance, including engagements with the European Union and the US-led Pax Silica initiative, where issues such as AI infrastructure, access to computing resources and responsible AI development are being discussed.
"We're part of various global fora looking at these issues and we've exchanged notes with countries and blocs including the EU and Pax Silica on what AI regulation could look like," Krishnan said.
The government has consistently maintained that India does not intend to replicate either the European Union's comprehensive AI Act or the relatively lighter regulatory approach initially adopted by the United States.
Instead, policymakers are working towards a framework that addresses the risks associated with frontier AI models while ensuring sufficient flexibility for startups, research institutions and innovation-led enterprises.
— IANS
Reader Comments
Finally, some sensible thinking from the IT Ministry. The US realizing they need regulation shows how fast things are changing. But I worry about implementation—our bureaucracy can be slow, and AI moves at lightning speed. We need dedicated AI regulators who understand tech, not generalists. Also, what about data privacy? The Digital Personal Data Protection Act has gaps. Hope AI regulation fills them.
Interesting to see India carving its own path. Having worked in AI regulation in the UK, I know how hard it is to balance innovation and safety. India's advantage is that it can learn from everyone's mistakes. The IndiaAI Mission sounds promising—hope it includes ethical AI research and not just commercial applications. Keep an eye on how it handles deepfakes and election integrity though.
A well-calibrated set of rules—that's what we need! As someone in a tech startup, I was worried about over-regulation killing our AI chatbot project. But this sounds like the government actually listened to industry stakeholders. Just hope the compliance burden doesn't become too heavy for smaller players. And please involve more developers and researchers in the consultation process, not just big corporations.
I'm cautiously optimistic. The EU's AI Act is comprehensive but may be too rigid; the US approach is flexible but risky. India's middle path could work if done right. My concern: will this actually be enforced? We have many laws on paper that aren't implemented effectively. Also, what about AI-generated misinformation in vernacular languages—that's a huge issue here that needs specific attention.
The IT Secretary seems to understand the
We welcome thoughtful discussions from our readers. Please keep comments respectful and on-topic.