From February 16 to 20, 2026, New Delhi, India, became the epicenter of one of the most important debates on the future of artificial intelligence: the AI Impact Summit. TEDIC was present thanks to the support of the Center for Communication Governance and NLU Delhi and participated in this summit, which brought together representatives from governments, international institutions, technology companies, and civil society organizations to exchange perspectives on how AI can be governed, regulated, and deployed on a global scale. To understand its scope and forms of impact, it is essential to examine recent developments from institutional and critical perspectives; through this article, TEDIC hopes to contribute to this multifaceted analysis of this agenda.
The AI Impact Summit, organized under the India AI Mission initiative, is part of a series of global summits on artificial intelligence that include previous events such as the AI Safety Summit (United Kingdom, 2023), the AI Seoul Summit (South Korea, 2024), and the AI Action Summit (Paris, 2025); following this last meeting, TEDIC began actively participating in this global initiative. In 2026, the event was held for the first time in a country of the Global South, raising expectations for a more inclusive shift in technology governance agendas.
The event’s official website describes the Summit as a global platform that fosters structured discussions on multiple dimensions of AI’s impact, including economic growth, social inclusion, governance, and technological innovation, featuring presentations, panels, forums, labs, and side events over the course of several days.
Despite the massive scale of the AI Impact Summit in India—with over 200,000 attendees, 530 panels, 50 delegations, and 20 heads of state including Emmanuel Macron, Lula da Silva, and the UN Secretary-General—some of the attention shifted toward tensions between corporate players and geopolitical signals from the tech sector. The presence of Google’s CEO, who arrived at the event after closing a multimillion-dollar deal with India, stood in contrast to the tense and widely discussed public greeting between the leaders of OpenAI and Anthropic, which highlighted the growing rivalry between the two organizations. At the same time, Nvidia’s absence—widely interpreted as a strategic reaction to the technology alliance between India and another country to invest more than $200 million in chip development—sent an implicit message about global competition in AI infrastructure and the race for leadership in advanced hardware.
Expectations and Pre-Summit Discussions
Civil society organizations focused on digital rights and international bodies raised several key expectations prior to the Summit’s opening:
1. A global agenda that gives voice to the Global South
For the first time, the AI Impact Summit was held in the Global South, which had fueled hopes that issues such as structural inequality, the distribution of benefits, and the socioeconomic risks of AI would be addressed in greater depth from the perspective of developing countries.
In this regard, it was hoped that the Summit’s agenda would facilitate mechanisms for greater participation by civil society and academia, leading to commitments that go beyond diplomatic rhetoric.
2. Coordination with international initiatives
Another expectation highlighted was the possibility that the event would align with existing global governance processes, such as those linked to the UN system like the Global Digital Compact, seeking to avoid fragmentation that intensifies asymmetries between core and peripheral countries. Or what was already done at BRIC 2025 regarding the Independent International Scientific Panel on AI and the AI Declaration, which recognizes the UN’s coordinating role on this issue.
What Actually Happened: A Multidimensional Landscape
1. International Participation and Economic Power
The Summit was a massive event: delegations from over 100 countries attended, including heads of state, ministers, CEOs of tech companies, and academic representatives. Major global AI platforms—including figures and leaders from companies such as Google, Microsoft, OpenAI, Meta, and other major corporations in the sector—used the event to forge alliances, present new initiatives, and announce investments.
The contrast in discourses was one of the event’s defining features. For example, while representatives from Brazil emphasized ethical dimensions, social justice, and regulatory responsibility. In the words of Lula da Silva, “when a few control algorithms and digital infrastructure, it is not about innovation, but about domination.” According to the Brazilian president, the global governance of artificial intelligence takes on a strategic role.
Other leaders focused on technological expansion, infrastructure investments, and global competitiveness. For example, India’s Modi made major announcements regarding large contracts with technology companies such as Google, Meta, Anthropic, OpenAI, and Microsoft on new investment plans for the Global South, including talent development, job training, infrastructure expansion such as undersea cables and data centers, support for small and medium-sized enterprises, and massive reskilling programs.
2. Large Scale of Activities
The sheer scale of the event made it difficult to keep up with the official agenda, which included hundreds of sessions, panels, innovation labs, technical debates, and side events. With over 200,000 attendees, 530 panels, and the distances between rooms and control centers, it was a significant challenge to make the most of the discussions and more political meetings.
The Formal Agreement: What Came Out of the Summit?
One of the most pressing questions was: What concrete agreements emerged from the AI Impact Summit? The main outcome was the Summit’s Final Declaration, signed by nearly 90 countries, including Paraguay. However, a critical point is that while this declaration establishes consensus on principles and guidelines, it is not legally binding in terms of international legal or regulatory obligations—a limitation also observed at the AI Summit in Paris.
The declaration organizes its vision around seven “chakra” pillars, including human capital development, trust, energy efficiency, democratization of resources, economic growth, science, and social empowerment. It also mentions voluntary cooperation platforms, such as the Charter for the Democratic Deployment of AI and the Trusted AI Commons.
However, from a critical perspective, this represents a convergence of ideas rather than a commitment to concrete action or accountability systems—an issue of concern to various stakeholders who had hoped for more binding governance advances.
Key tensions identified
1. Inclusion vs. Techno-Optimism:
Digital rights organizations and academics have pointed out that, although the official declaration aims to include diverse voices, in practice structural discussions on social risks—such as algorithmic bias, data sovereignty, or labor impacts—have been subordinated to narratives that celebrate technological innovation as a driver of growth.
2. Human Rights Take a Back Seat:
A point highlighted by both Data Privacy Brasil and critical analyses is that the text of the declaration and much of the discussion do not explicitly incorporate human rights-based protection mechanisms**, such as the right to privacy, non-discrimination, or social justice.
3. Role of Civil Society:
Although the Summit was promoted as a “multistakeholder” space, the participation of civil society and academic groups in decision-making forums was limited: few formal mechanisms ensured that their contributions directly informed the final texts or agreements, a recurring criticism in post-event analyses.
Post-Event Resources
Despite these tensions, the Summit produced some useful resources: casebooks on AI projects addressing social issues, available on the event’s official website (*see the casebooks section*). Working group (WG) documents containing operational proposals and guidelines, such as a platform for sharing computing resources among countries or principles for training workers in the age of AI. These materials can serve as a reference for public institutions, academic institutions, and organizations seeking to translate global discussions into public policies and technical practices.
What’s Next for AI Governance
The next AI Summit is scheduled for 2027 in Switzerland, according to an announcement by Swiss Ambassador Thomas Schneider in India. His remarks struck a more optimistic tone regarding the inclusion of various stakeholders and coordination with existing initiatives in other multilateral forums. Switzerland also signaled a greater willingness to advance binding instruments, such as the Council of Europe’s Vilnius Convention on Artificial Intelligence, Human Rights, Democracy, and the Rule of Law—the first legally binding international treaty in this area. It is hoped that the Convention can be extended to other countries, fostering a shared and interoperable vision without hindering coexistence with national regulatory frameworks.
Through the Global South Alliance (GSA), of which TEDIC is a member, the organization has been involved since December 2025 in discussions leading up to and parallel to the AI Summit in India. In the days following the event, TEDIC continues to actively participate in meetings to follow up on the agenda through the MAP AI, led by Indian members within the Global Network Initiative (GNI).
With 2026 still in its early stages and amid expectations surrounding new AI governance frameworks—such as the UN’s Independent Scientific Panel and the Global Dialogue—there remains hope that these processes will help strengthen the currently fragile multilateral system and enable effective multisectoral participation on a more equitable basis, where human rights, and not just industrial policy, take center stage. In the face of resistance from major countries, such as the United States, to more robust frameworks for global governance within the United Nations, the challenge remains. In this context, expectations are growing regarding countries like India, Brazil, and Switzerland, which, although they do not hold significant power in AI, seek to strengthen governance and regulatory structures capable of reducing asymmetries.
Defending those who defend us