The recent US restrictions on advanced AI technologies have reignited a familiar question for India. Can the country build a truly sovereign AI ecosystem, or will it remain dependent on foreign models and infrastructure?

The debate has gained urgency as AI increasingly becomes a strategic technology akin to cybersecurity, semiconductors and telecommunications. While the government's IndiaAI Mission has committed substantial resources to support domestic AI development, industry leaders argue that achieving genuine sovereignty will require far more than funding announcements.
Business Today spoke to startups selected under the IndiaAI Mission to understand what India has achieved so far, the gaps that still remain, and what it will take to build a genuinely sovereign AI ecosystem.

"There are three key aspects to building frontier models—people, data and compute," says Ganesh Gopalan, CEO of AI startup Gnani.ai, which raised $10 million in a Series B round earlier this year led by Aavishkaar Capital. According to Gopalan, while investments are beginning to address compute infrastructure and data availability, India's biggest advantage lies in talent.

MUST READ: Sarvam's Vivek Raghavan on why sovereign AI is no longer optional

"The best opportunity in India lies in AI talent, which will be the biggest differentiator in the future, and India has a traditional advantage," he says. Gopalan adds that AI sovereignty is becoming "non-negotiable" not just for countries but also for enterprises, with government initiatives such as the IndiaAI Mission and the proposed Research, Development and Innovation (RDI) Fund helping lay the groundwork.

Yet industry executives acknowledge that India remains significantly behind global leaders. "Both the United States and China are way ahead in this game," says Vivek Raghavan, co-founder of Sarvam, India's homegrown AI startup that recently raised $234 million in one of the largest funding rounds in the country's AI sector. The funding, he says, reflects growing market confidence that Indian companies can build foundational AI capabilities despite the gap.

For Raghavan, sovereignty is not optional. "Even if we are lagging behind, we have to do it," he says, arguing that AI capabilities will increasingly underpin national security, cybersecurity and digital infrastructure. He believes India's focus should be on developing models and agentic AI systems tailored to Indian enterprises, government use cases and citizen services.

Suraj Amonkar, Chief AI Research Officer at Fractal, one of India's earliest AI companies to go public says the rapid progress of agentic AI demonstrates both the opportunities and risks of dependence on external technologies.

MUST READ: Here's why HCLTech is betting $150 million on Sarvam

"The subsequent restrictions placed on advanced models also demonstrate the importance of sovereignty considerations for the most advanced AI systems," Amonkar says. He believes India must simultaneously pursue frontier AI research while becoming a "use-case capital" where AI adoption scales across industries.

The challenge, however, remains formidable. While India is building compute capacity, funding domestic startups and nurturing talent, experts say true AI sovereignty will require sustained investments at a scale comparable to global competitors, alongside long-term commitments to indigenous models, infrastructure and research.

MUST READ: India accelerates AI development by backing 12 AI projects after Anthropic Fable 5 ban