That AI Safety Summit, steeped in moral urgency, raised an important alarm: Artificial intelligence (AI) may be the most powerful technology humanity has created and we have no consensus on how to govern it.
Also Read: The Paris AI Action Summit could help India raise its game in this vital field
Six months later, the conversation moved to Paris for the AI Action Summit. The mood had shifted. Less philosophy, more policy. Countries debated surveillance versus safety, openness versus sovereignty. The EU showcased its regulatory strides; others offered national strategies. The US and UK demurred from the evolving consensus. From ethics to execution, the message was clear: the world was waking up, but still speaking in different tongues.
Now, the baton passes to India. In February 2026, New Delhi will host the AI Impact Summit, the third major global checkpoint in the ongoing effort to define AI’s future. Hopefully, this is more than just another gathering of talking heads, but a rare opportunity for India to lead the conversation past alarm and action into actual impact.
The world is at an inflection point. Generative AI is no longer a novelty. It writes code, composes music, conducts research, automates jobs and even drafts legislation. In boardrooms, classrooms and courtrooms, AI is becoming invisible infrastructure, like electricity. The US is moving towards a hands-off regime, China’s approach is state directed. Europe is building a framework of rules. Most countries, including India, are still watching from the sidelines, unsure where to stand.
Also Read: Outrage over AI is pointless if we’re clueless about AI models
By 2026, we will likely have seen new versions of GPT, Gemini and Claude, which will be even more powerful, agentic and embedded. The guard-rails we set now will shape not just product design, but societal outcomes. We have to decide whether AI will amplify inequity or bridge it, empower individuals or surveil them, and will it accelerate development or deepen divides. India’s summit, then, is not just another event. It’s a platform from which a new kind of global AI diplomacy can be launched.
The AI conversation has so far been dominated by corporations and researchers, but it is governments that represent the will of the people and must create enabling environments for both safety and innovation. Too often, however, intergovernmental summits end up as echo chambers of good intentions and empty communiqués.
What we need now is some refreshing practicality: not just talking about AI governance, but showing it as the word ‘impact’ suggests. The AI Impact Summit must push for governments to open-source their playbooks—frameworks for data sharing, national compute strategies, skilling programmes and regulatory sandboxes. What worked in Europe may not work in Ethiopia, but all learnings must travel. India should invite countries to present not just vision statements, but detailed ‘AI governance manifestos’: what they are doing on AI ethics, safety, infrastructure, talent and societal impact. A kind of AI CoP, if you will.
India knows how to host world-class events like the G20 in 2023. Arguably, though, the outcome of that summit was overshadowed by its spectacle.The AI Impact Summit must avoid that trap and India should be aiming for impact and outcome over performative wins. One idea is to design inbuilt accountability. Every commitment made at the summit should be assigned a lead country, a timeline and a review mechanism. Let the summit lead to a dashboard rather than just a declaration.
Also Read: Leaders, watch out: AI chatbots are the yes-men of modern life
India is uniquely positioned to do this. It is the world’s largest democracy, a tech hub of startups and corporates, and holds a leadership position in the Global South. Its world-leading efforts with digital public infrastructure (DPI) have shown that inclusive tech at scale is possible. India can bring its DPI playbook to the global table with AI. India also has moral legitimacy. It can speak credibly on data dignity, avoiding AI colonialism and ensuring that foundational models are trained on world-views that are not just Western. It can advocate a shared ‘AI Commons,’ a pool of datasets, models and compute power that’s open, multilingual and inclusive.
This is also an opportunity for the world’s biggest tech companies and labs to co-create the agenda. The summit must insist on concrete commitments from them: sharing evaluation benchmarks, collaborating with governments on safety testing, investing in compute power for public-good AI and publishing transparency reports on model training and usage.
The year 2026 could see India take leadership of defining the desired impact of AI—AI for Good and AI for All. If India pioneers this effort, the site of its summit could come to be known as the Bretton Woods of AI: where ethics meets execution, safety meets scale and AI meets humanity.
The author is a founder of AI&Beyond and the author of ‘The Tech Whisperer’.
