India debates AI future as Anthropic closes access to new models

Anthropic’s sudden move to cut off access to its latest AI models in accordance with U.S. government guidelines has raised new questions across the global technology industry. In India, the decision has reignited a long-running debate over whether one of the world’s largest AI markets can rely on technology built and controlled elsewhere.

The announcement comes as Anthropic said late Friday that it had been directed by the U.S. government to suspend access to all foreign nationals, including its foreign employees, to its recently launched Fable 5 and Mythos 5 models. The move comes shortly after the company announced a partnership with Indian IT services giant Tata Consultancy Services to expand enterprise AI adoption in India, highlighting how closely India’s AI ambitions are tied to technologies developed and managed in the United States.

The broader implications are unclear, but some reports say the initial security concerns were first reported to the government by Amazon CEO Andy Jassy. And The Information said the White House is unlikely to extend similar restrictions to other AI companies and has privately criticized Anthropic’s handling of the jailbreak vulnerability. Anthropic disputed the character of the government and argued that such action should not have been taken.

Nonetheless, the development has sparked a debate among Indian founders, investors and policy experts over whether India should accelerate its efforts to build domestic AI capabilities, deepen its investments in open source alternatives, or continue to rely on a small number of pioneering US model providers. For some, this episode is a wake-up call about technology dependence. For others, it’s a reminder that access to increasingly important AI systems may be shaped by geopolitical decisions beyond India’s control.

India has become one of the most important markets for advanced AI companies. Anthropic and OpenAI described the South Asian country as the second-largest market after the United States, reflecting its growing importance in the global AI race. These companies have already established offices in India and have expanded local hiring, partnerships and corporate initiatives in recent months, leveraging India’s broad base of developers, startups and enterprises to accelerate adoption of the latest technologies.

For many in the Indian tech space, Anthropic’s announcement on Friday was about more than just one AI company. This has again raised questions about the country’s long-term AI strategy and whether India can afford to continue relying on a small number of foreign AI providers.

“This completely changes the situation,” said Aakrit Vaish, founder of Indian AI venture platform Activate, referring to Anthropic’s decision. “I think this really changes the way we all need to think about sovereign AI in India.”

Vaish told TechCrunch he was “shocked and confused” upon hearing the announcement on Saturday morning and said it strengthened the case for developing domestic AI capabilities. He expects startups to increasingly shift to open source models and plans to encourage companies in his portfolio to reduce their reliance on a small number of leading AI providers.

For some founders, a bigger concern was how limiting access to cutting-edge AI could affect their competitiveness. Vijay Rayapati, co-founder and CEO of Atomicwork, told TechCrunch that the episode highlights the risks startups with teams spanning multiple countries face if access to cutting-edge AI systems becomes subject to geopolitical restrictions.

Atomicwork has about 25 employees in the U.S., but most of its product engineering team is based in Bengaluru, India.

“If your AI team is not comprised entirely of U.S. citizens, you are at a competitive disadvantage,” Rayapati said. He argued that unequal access to cutting-edge AI models could give some companies a significant advantage over their competitors.

These concerns come as parts of India’s tech sector are already grappling with questions about how AI could reshape the global talent economy. This week, American real estate technology company Opendoor closed its India office, less than two years after expanding in the country. CEO Kaz Nejatian cited efforts to bring operations closer to U.S. customers and a shift to smaller, AI-driven teams.

Opendoor did not specify how many of its decisions were driven by AI-related efficiencies, but the move sparked a broader debate about how advances in AI could impact the future of global tech work, and what this could mean for India’s position as an engineering talent hub.

beyond humanity

Beyond startups and AI builders, the Anthropic episode has sparked widespread debate among India’s tech leaders about its dependence on foreign AI infrastructure.

Sridhar Vembu, founder of Indian SaaS company Zoho, said the move showed “technology is the ultimate weapon” and urged Indian organizations to increasingly embrace open source models on a smaller scale.

Vembu wrote to

Investor and former Infosys executive Mohandas Pai, responding to Vembu on

“We are far behind and we need a national mission to advance quickly,” Pai said, urging the government to create an annual 500 billion rupees (about $5 billion) fund for AI and deep technologies, along with a 2 trillion rupees (about $21 billion) credit guarantee program to support cloud infrastructure, hardware and semiconductor development.

Pai’s proposal would scale back India’s existing AI efforts. In 2024, New Delhi approved the IndiaAI mission, spending 103.72 billion rupees (about $1.2 billion) over five years, aimed at expanding computing infrastructure, supporting startups and developing indigenous AI capabilities.

Despite growing interest in AI and New Delhi’s efforts to develop domestic capabilities, India still plays a relatively small role in frontier model development. Only a few startups are pursuing foundational AI models, including Sarvam, which launched an open source model earlier this year. However, Krutrim, another popular AI startup, initially positioned itself around foundational model development and then shifted to cloud and AI infrastructure services.

Instead, much of the Indian AI ecosystem has focused on applications and specialized models built on top of existing foundational models. A recent example is Avataar AI, which launched a video generation model earlier this week to provide a cheaper alternative to products from competitors including Google’s Veo, Kling, Luma and Runway.

Not everyone agrees that the biggest problem is lack of capital. In response to Pai’s comments, Lightspeed partner Hemant Mohapatra argued that the biggest constraint to building a globally competitive AI company is not simply the size of the investment commitment, but talent, access to computing resources, and execution.

Mohapatra estimated that training a cutting-edge AI model can cost anywhere from hundreds of millions to billions of dollars, depending on the approach, but said successful AI companies have historically expanded their capital requirements over time as adoption increases.

But for some policy observers, the implications extend beyond AI startups or model providers.

Prasanto Roy, a New Delhi-based technology policy expert who advises multinational companies, said the episode would reinforce concerns within the Indian government about its strategic autonomy. This compares to the lessons learned from Russia’s loss of access to SWIFT and other parts of the global financial system following its invasion of Ukraine.

He told TechCrunch that the move was likely to spark a significant nationalist backlash in India, describing it as a poorly considered decision by Washington, with ramifications extending far beyond Anthropic itself.

“Even if this is modified or reversed, the humanity episode shows that there is no such thing as a geopolitically neutral foreign LLM,” Roy said. “The U.S. AI model is tied to U.S. geopolitics.”