As global competition in artificial intelligence intensifies, the race is shifting. It is no longer just about building the most powerful chip or the largest data centre. According to Samir Saran, President of the Observer Research Foundation, writing on December 24, 2025, the true victors will be those who can seamlessly integrate AI into the fabric of daily life on a national or continental scale. For India, this presents a monumental opportunity to leapfrog in the new technological age.
The Three Eras of AI: Compute, Diffusion, and Governance
Saran outlines a framework of three distinct phases defining the AI age. The first is the Compute Era, dominated by the scramble for semiconductor supply chains and hyperscale data centres. This phase favours entities with massive capital, energy, and traditional resources like land and water.
We are now witnessing the potential peak of this era, with signs of overinvestment in physical infrastructure. Computational power is on track to become a regulated utility, much like electricity or water. The real value, however, will be created in the subsequent Diffusion Era.
This second phase will be led by nations that can deploy AI at population scale across their real economies. Power will flow not only to algorithm owners but to those who weave AI into everyday life. Countries like India, with proven expertise in adopting and adapting technology for massive, diverse populations, hold a natural advantage here.
The final phase is the Governance Era, where sovereign states will assert control, guarding data sovereignty and shaping the competitive landscape through regulation and geopolitical leverage.
India's Crucible: The Three Critical Questions
To dominate this threefold AI journey, Saran argues that India must find convincing answers to three pivotal questions.
First, can Indian companies take the necessary risks? While agile startups have shown entrepreneurial flair, the broader private sector must cultivate a new mindset. Embracing the Diffusion Era requires a greater appetite for risk, changes in corporate governance, and the courage to bet on an uncertain technological future.
Second, can Indian institutions mobilise patient capital? The nation needs financial systems capable of funding long-term, utility-scale infrastructure and betting on undiscovered innovations. The government's role is crucial in crafting a regulatory model that balances private profit with public good, akin to the successful Digital Public Infrastructure (DPI) approach. This framework must protect individual rights and digital sovereignty without extinguishing entrepreneurial energy.
Third, can India build a unique AI architecture? The nation must develop an AI ecosystem that sits comfortably within global systems yet addresses local contexts. Current large language models (LLMs) are often trained on datasets that over-represent Western perspectives, misrepresenting cultures like India's. The Diffusion Era demands localization, where new winners and solutions emerge from within communities, fostering trust and perceived value.
The Path Forward for a Leading AI Nation
The transition from the Compute Era to the Diffusion Era is India's moment to shine. The blueprint for success involves a triad of actions: building robust compute infrastructure, unleashing the nation's entrepreneurial communities, and creating supportive, forward-looking regulation.
If India can successfully answer these three questions—fostering corporate risk-taking, channelling patient capital, and establishing visionary governance—it is poised to not just participate in the AI revolution but to lead it through the coming Diffusion and Governance eras. The opportunity extends far beyond hardware; it lies in harnessing AI for inclusive, large-scale impact.
