Training advanced AI models requires enormous computing power, specialised chips, and massive datasets
| Photo Credit:
BlackJack3D
In the previous column we discussed how India risks being left out of AI technology ownership. However, all is not lost and the country can still take steps to ensure we don’t end up merely as an AI consumer.Indian engineers run global technology companies, startups in India attract billions in investment and our IT exports are the envy of nations that recognise us as a technology superpower, but all is not perfect. Instead of owning technology, we have become the world’s digital workforce. As the world stands on the cusp of an incredible revolution, it is important to understand that AI isn’t just another technology but part of a nation’s strategic infrastructure, just like energy, factories, defence capabilities or telecommunication networks; hence it will be subject to sudden restrictions, as happened recently with the US announcement on Anthropic’s advanced models.Startups like Sarvam are trying to build indigenous capabilities instead of developing wrappers around overseas foundational models. But building frontier AI is not like launching an e-commerce company or a SaaS platform. It requires a broader ecosystem, and the economics are fundamentally different. Training advanced models requires enormous computing power, specialised chips, and massive datasets, and the cost can run into billions of dollars.A meaningful AI strategy could require ten years of massive investment before we can expect outcomes at global level. Private investors may find this uncomfortable but the government cannot. Apart from world-class computing infrastructure, advanced chips, research funding and close collaboration between universities and industry, India needs patience. We cannot make the mistake of comparing every Indian effort with OpenAI or Anthropic.The objective should be sovereignty, not supremacy. Critical capabilities must remain available to us, regardless of geopolitical developments. As India embarks on this AI journey, the risks are obvious. Large investments may fail badly. Technology may evolve faster than expected and global leaders could continue to extend their lead. But the risk of inaction is greater.For decades, technology globalisation allowed India to access the best tools in the world. However, the future will belong to countries that own critical technologies. India has spent three decades supplying the world’s tech talent, but the next three decades will be determined by whether it can own the technologies that shape the world. The question is no longer whether India can afford to invest in AI, but whether it can afford not to.(The writer is a serial entrepreneur and best-selling author of the book ‘Failing to Succeed’; posts on X @vaitheek)Published on July 6, 2026
