India Negotiates with US for Access to Anthropic's "Project Glasswing"

en.Wedoany.com Reported - S Krishnan, Secretary of India's Ministry of Electronics and Information Technology, stated that India is negotiating with the United States to gain access to Anthropic's "Project Glasswing" for stress-testing critical digital infrastructure, even though India has already begun testing its systems using existing AI models.

Speaking at the Confederation of Indian Industry's Cybersecurity Summit, Krishnan revealed that access to Anthropic's product would help India test its systems' ability to withstand advanced AI capabilities. He stated that he is negotiating with his US counterparts to obtain access to "Project Glasswing," emphasizing that this would facilitate stress-testing India's own systems on Anthropic's platform. However, India is not waiting for the outcome of these negotiations. Krishnan noted that the country already has the capability to test critical code using existing models and should complete 60% to 70% of the testing work with these tools while continuing negotiations with the US. This statement echoes the earlier view of NASSCOM Chairman Srikanth Velamakanni, who argued that India should continue seeking access to the world's most advanced proprietary AI models while simultaneously using increasingly powerful open-source alternatives to probe vulnerabilities in its own infrastructure. Velamakanni cited models such as Kimi K2.7, GLM 5.2, and Qwen 3.7, asserting that India should not rely on a few proprietary frontier models for cybersecurity testing.