Optimism, tempered with a note of caution, marked Prime Minister Narendra Modi’s opening address at the AI Action Summit in Paris on Tuesday. Co-chairing the event with French President Emmanuel Macron, the Prime Minister spoke of the “absolutely amazing” positive potential of AI in health, education, agriculture and other sectors, while also underlining the need to be alert to its potential for “biases”. As he did so, he outlined a vision for the future where “open-source systems…enhance trust and transparency”, where technology is democratised and “rooted in local ecosystems” — words that carry a special resonance in the post-DeepSeek world, where old certainties about capital- and compute-intensive models built in a closed ecosystem have been shaken by the open-source, cheaply made Chinese model.
The Prime Minister’s pitch for the open-source route comes in the context of India’s own growing ambitions on the AI front. It hints at the desire to strike out on a path that diverges from the one that has, so far, been laid by Silicon Valley behemoths like Google and Microsoft. It signals India’s aspiration to be more than just a market for models developed elsewhere, such as the ones by OpenAI, whose CEO, Sam Altman, showed much excitement for the country’s vast, and growing, user base on his visit last week. India, as the PM noted, is pushing to build its own large language model (LLM), and ensure a place for itself in a contest that has intensified in recent months. How intense, and indeed fraught, this race has become, was evident at the summit, with two participants, the US and the UK, refusing to sign the summit declaration — the former on the grounds that excessive regulation could stifle innovation and challenge its domination of the sector, and the latter because the statement did not go far enough in addressing global governance of the technology. Finding common ground between such competing visions and working towards a global concerted approach to AI is the challenge going forward.
