India should continue engaging with the US government for access to advanced AI models while aggressively testing its critical infrastructure using powerful open-source alternatives, according to Nasscom Chairperson Srikanth Velamakanni. He said India can download recently released open-source, open-weight models such as Kimi K2.7, GLM 5.2 and Qwen 3.7 to identify vulnerabilities across financial and other critical infrastructure. His comments follow tighter US export controls on frontier AI technologies. Velamakanni added that India should simultaneously accelerate investments in its own foundation models. Regarding multi-billion-dollar investments by Microsoft, Amazon and Google into India's AI infrastructure, he said these will create jobs, spur ancillary industries and improve consumer experience by reducing latency. On concerns about AI hurting India's $280-billion IT services industry, he cited Jevons paradox, stating cheaper AI would expand software usage and create fresh demand for technology talent, adding that "the world will need a thousand times more software than it has today."