FRENCH CAFÉ MEETS DESI CHAI: Recipe from Paris to Delhi for Democratic AI

Picture this: You’re sitting at a café in Paris, sunlight spilling onto the nappe, having a conversation about whether Artificial Intelligence (AI) ethics should be applicable to the recipes of croissants & coffee. Now, imagine having the same conversation at a tea stall in Delhi, where the amount of sugar in every cup, just like AI regulation, is subjected to comprehensive debate & deliberation. Who knew that AI could blend into our daily lives as a part of our life as much as our morning beverage?

AI has become a bugbear in global policy discourse. The entire comity of nations stands at the cusp of technological transformation without knowing how to respond to it. As AI advances at espresso-shot speed, several questions arise for nations to address unanimously- who will set the rules for AI development and usage? Who will benefit, and who will lose as AI penetrates further into our lives? And most importantly, will AI be seen as a collective good that binds nation states, much like space technology and IVF technology, or will it deepen the North-South divide like nuclear technology?

Paris AI Summit 2025: From Existentialism to Algorithms:

Earlier this year, the proponents of Liberty-Egality-Fraternity hosted the Paris AI Summit which highlighted the challenges for policymakers in accounting for context-specific capabilities in AI regulation. Since technical models are based on continuous fine-tuning and code enhancements, any policy would face the risk of running its course in a short span of time. After all, advancements in technology move at an unprecedented rate- hard to catch up, arduous to regulate.

The Summit received a collective nod towards the identification of existing risks (International AI Security Report), creation of AI Public Goods (Current AI) and acceptance of building a sustainable AI system (Environmental Sustainability Coalition). However, just as no two croissants taste the same, the two parts of the world weren’t in consensus. The developing countries raised questions regarding the what-fits-America-fits-all approach of the Summit. For nations grappling with supply chain disruptions due to geopolitical tensions, subpar infrastructure for technology demonstrations, and varied socio-economic priorities, the question remained: how would AI regulations pan out?

India AI Summit 2026: Chai Under Construction:

As India is set to host the next edition of AI Summit, there is already an environment of strong opinions and towering expectations. After all, India boasts itself on three relevant grounds- voice of the global south, the largest and the most successful democracy in the world, and having the world’s largest digital population contributing substantially to data & content.

The Summit will focus on three ‘sutras’- people, planet & progress. These would further permeate into seven ‘chakras’ (Human Capital, Inclusion, Safe & Trusted AI, Resilience, Science, Democratizing AI Resources & Social good)- the concrete areas of action.

If expectations are to be accounted for, the Global South is counting on India to present a united and democratic front such that developing nations not only receive the resources for capacity building, but are also able to retain sovereignty with respect to the regulations. The resultant AI Policy would then exhibit flavours of democracy, inclusivity and interoperability. Dependency on external actors without checks & balances can leave developing countries vulnerable. Therefore, tech-diplomacy seems to be an unequivocal choice to resolve the challenges of technology competitiveness.

However, with big talks and no action, the discourse becomes redundant. At the end of the day, technology is not divided by borders. So, why should we?

French Café Meets Delhi Chai:

The overarching objective of framing a democratic regulatory framework should be to ensure that every decision and every discussion on the usage, development, profit distribution & governance of AI is reflective of the interests & aspirations of all the countries. With Delhi leading the ‘Impact’ phase of the summit, it can take into consideration the following recommendations.

First, the creation of the AI Public Goods is inessential without comprehensive data. Private players might not be able to create high-quality authenticated datasets. This is where the Global South comes in. With 85% of the world’s population and concentration of cultural industries in the area, New Delhi could facilitate the creation of these goods. While the North and private players bring in money for dataset development, the South brings in the spice!

Second, the South needs to understand that merely adapting to the models laid out by the developed countries would enmesh them in a vicious cycle of catching up and digital colonialism. The technological imbalance between the North and the South is not just economic, rather it shapes AI technology. Why should the AI safety standards formulated in Brussels decide how the artisans of Rajasthan should deal with it? Thus, the aim is to become active contributors to the evolution of AI, rather than mere markets, tech-dumps and data mines.

Third, India should ensure that AI is kaldor-hicks efficient, where even if the interests of a section of the society are compromised, they are fairly compensated through adjustment assistance. This could be done on a country by country basis, where nations roll out domestic guidelines to reskill & upskill the population prone to future job losses and income stagnation. Cues can be taken from United Kingdom’s National Retraining Scheme 2017 and Amazon’s Upskilling Program 2025.

Fourth, India needs to percolate the principles of deliberative democracy within the entire mechanism. At the global level, it could ask every country to come up with a single demand/solution to the menace created by AI that can be included in the framework. The solution could be as local as Jugalbandi, a generative AI-driven chatbot on mobile devices which provides information about government schemes in regional languages.

Fifth, India needs to step away from the elite model of policy making. Consultations with various stakeholders would only be fruitful, if they are provided with prior information about the vision of the policy to be formulated. The walls between the strategic community and the government need to come down for clear and resonating impact. As soon as these walls come down, India can move towards a comprehensive policy that places India as a responsible leader on the global map. At the same time, India must be patient enough to let these various ingredients boil to attain an exquisite taste.

Brewing futures beyond borders:

A perfect and definitive policy cannot be brewed up in a single summit. The process could be slow, but it needs to be robust. If Paris laid down the ingredients for us, it is India’s responsibility to test the recipe- one where technology is about diversity, inclusivity & humanity. The conversation should not be about which part of the world is more capable of laying out AI Principles, but about how the tech-superpower nations can collaborate with aspirational nations to bring about tangible & practical solutions. After all, the future of AI might be written in code, but its flavour will surely belong to those who have the sense to cooperate, collaborate and brew solutions together defying borders.