Novel Delhi’s Bharat Mandapam recently served as the epicenter for a global debate on the future of machine intelligence during the AI Impact Summit 2026. Held from February 16 to 21, the event marked a pivotal shift in the international approach to artificial intelligence, moving away from the theoretical safety frameworks of previous years toward a focus on practical implementation and measurable outcomes.
The summit was the first in its series to be hosted by a Global South nation, signaling India’s ambition to lead the narrative on how AI can be deployed for social good and economic growth. Organized under the IndiaAI Mission by the Ministry of Electronics and Information Technology, the gathering brought together over 20 heads of state, 60 ministers, and the architects of the world’s most powerful AI systems, including Sam Altman of OpenAI and Sundar Pichai of Google.
While the event was designed as a showcase of “frugal, sovereign and scalable” innovation, it was not without friction. The week was characterized by a stark contrast between high-level diplomatic success—highlighted by the presence of French President Emmanuel Macron and UN Secretary-General António Guterres—and a series of organizational lapses and political protests that underscored the complexities of hosting a high-security global forum in the Indian capital.
The transition from the “Safety” and “Action” themes of previous summits in Bletchley Park, Seoul, and Paris to “Impact” reflects a broader geopolitical trend. According to analysis from Crowell & Moring, the shift suggests that global powers are now less concerned with if AI can be governed and more concerned with who will control the infrastructure and how it will be monetized in lower-income regions.
Sovereign Ambitions and Infrastructure Pledges
At the heart of the summit was India’s push for “sovereign AI”—the idea that a nation should possess its own compute capacity and data models to avoid dependence on foreign technology. Union Minister Ashwini Vaishnaw detailed a “whole-of-nation” strategy to build an ecosystem that is both scalable and cost-effective. A key component of this plan is the expansion of the IndiaAI Compute Portal, with the government announcing plans to add more than 20,000 GPUs to the existing base of 38,000.
The private sector also made significant commitments to bridge the digital divide. Microsoft announced We see on track to invest $50 billion by the end of the decade specifically to bring AI capabilities to lower-income countries. These infrastructure plays were complemented by the unveiling of homegrown models, including the government-backed BharatGen Param2, a 17-billion parameter model supporting 22 Indian languages.
The summit’s conceptual framework was built around three “Sutras”—People, Planet, and Progress. These pillars guided seven thematic working groups tasked with delivering outcomes in areas ranging from human capital and science to the democratization of AI resources. The scale of public interest was evident when the government extended the event by one day, moving the conclusion to February 21 to accommodate an overwhelming response from students and tech enthusiasts.
Key Technological Debuts
- Sarvam AI: The Indian lab launched a new generation of large language models, including 30-billion and 105-billion parameter versions utilizing a mixture-of-experts architecture.
- Kaze Smartglasses: Sarvam’s first hardware product was tested by Prime Minister Narendra Modi during the expo.
- BharatGen Param2: A multimodal, 17-billion parameter model designed for linguistic inclusivity across India.
A Record of Responsibility and its Critics
In a bid to project a commitment to ethical AI, India partnered with Intel India to launch an AI responsibility campaign. The effort resulted in a Guinness World Record, collecting 250,946 valid pledges within 24 hours, far exceeding the initial target of 5,000.
However, the summit’s image of seamless innovation was punctured by a high-profile incident involving Galgotias University. A representative at the university’s pavilion presented a robot dog as an indigenous development, only for social media users to identify it as a Unitree Go2, a commercially available product from the Chinese company Unitree Robotics. The embarrassment was compounded when it was revealed that Union IT Minister Ashwini Vaishnaw had shared a video of the robot before deleting it. IT Secretary S. Krishnan subsequently directed the university to vacate its stall, and the institution issued an apology, citing an “ill-informed” representative.
Beyond the “robot dog” controversy, the event faced scrutiny over its structure. TechPolicy.Press argued that the summit granted multinational corporations parity with sovereign governments via the CEO Roundtable, while offering no equivalent high-level platform for civil society or human rights defenders. Further reports suggested that the U.S. Delegation viewed the event not as a cooperative effort, but as a geopolitical race for domination against China.
Logistical Chaos and Political Friction
The prestige of the Bharat Mandapam venue was occasionally overshadowed by organizational failures. Bloomberg reported that delegates were left stranded without food or water during a security lockdown on February 19 ahead of the Prime Minister’s visit. This lockdown also meant the venue was closed to the general public on that day, sparking frustration among registered attendees.
Security concerns were further highlighted by an incident involving Dhananjay Yadav, a Bengaluru-based entrepreneur, who alleged that his wearable products were stolen from within the high-security zone. While the Delhi Police eventually recovered the devices, the incident raised questions about the efficacy of the summit’s security protocols.
Think about this: We paid for flights, accommodation, logistics and even the booth. Only to see our wearables disappear inside a high-security zone
The political climate in Delhi also spilled into the summit. On February 20, members of the Indian Youth Congress (IYC) staged protests inside the venue, criticizing a recent trade deal between India and the U.S. And shouting slogans targeting the Prime Minister. Four protesters were sent to police custody on February 22, though they were later granted bail by the Delhi High Court on March 2, which described the protest as a “political critique.”
| Date | Event | Key Outcome/Incident |
|---|---|---|
| Feb 16 | Expo Inauguration | 300+ exhibitors from 30 countries launch pavilions. |
| Feb 18 | Research Symposium | IIIT Hyderabad leads talks on sovereign AI infrastructure. |
| Feb 19 | Leaders’ Plenary | PM Modi, Pres. Macron, and Sec-Gen Guterres address the summit. |
| Feb 20 | IYC Protests | Internal protests lead to police custody of four members. |
| Feb 21 | Closing Day | Summit extended by one day due to public demand. |
As the dust settles in New Delhi, the focus now shifts toward the next iteration of this global dialogue. The series is scheduled to continue in 2027, with Switzerland set to host the next AI Summit in Geneva, potentially shifting the conversation back toward the governance and safety frameworks that defined the early meetings in the UK and South Korea.
We invite our readers to share their perspectives on the balance between sovereign AI ambitions and global cooperation in the comments below.
