New Delhi concluded a five-day global summit on artificial intelligence on , bringing together heads of state, technology executives, and policymakers to address the rapidly evolving landscape of AI development and governance. While the India AI Impact Summit resulted in a significant investment pledge and a non-binding declaration of principles, observers noted a gap between ambition and concrete commitments, particularly regarding the concentration of power within a handful of nations and corporations.
The summit, the fourth in a series following gatherings in the U.K., South Korea, and France, marked the first time such a forum was hosted in the Global South. A key objective for many attendees was to establish a pathway for middle powers to influence the direction of AI, ensuring its benefits are not solely captured by the United States and China. The event, however, was described as “thinner on binding commitments” and left some hoping for a more substantial shift in global AI governance with “mixed feelings.”
The New Delhi Declaration
The primary outcome of the summit was the adoption of the New Delhi Declaration on AI Impact by 88 countries and international organizations. The declaration outlines principles for inclusive and human-centric AI development, with ambitions to democratize access, expand AI’s application in sectors like healthcare and education, and ensure ethical safeguards and transparency. Both the United States and China endorsed the declaration, signaling a degree of consensus on broad principles.
Despite the widespread endorsement, the declaration’s practical implications remain limited. It acknowledges the existing imbalance in access to the resources – computing power, data, and expertise – required to develop cutting-edge AI models, but does not offer concrete solutions. The operational details of the declaration are also notably sparse.
A central concern voiced throughout the summit was the potential for AI to exacerbate existing power imbalances. A significant portion of global AI development is currently driven by a small number of American companies, whose proprietary models and infrastructure dominate the field. China represents the other major force, controlling roughly 90% of global AI computing infrastructure alongside the US. While efforts are underway to develop alternative, open-source models, these currently lack the scale and sophistication of their American and Chinese counterparts.
Mark Surman, president of Mozilla, highlighted a desire among many participants for a more open and sovereign approach to AI. “But beyond the cameras we saw real hunger from countries, companies and communities to come together and build AI that is open source, sovereign and culturally tailored,” he said.
In Europe, anxieties surrounding reliance on American technology have been heightened by recent geopolitical events. Seán Ó hÉigeartaigh, director of AI FAR at the University of Cambridge, observed that a perceived lack of understanding in Washington regarding European concerns has complicated efforts to forge collaborative strategies. “Feels like they’re still reading from last year’s notes. Trying to push positions and strategies that will no longer work,” he stated.
Arthur Mensch, CEO of French AI company Mistral, argued for greater autonomy in AI infrastructure, asserting that organizations – governments, hospitals, and public institutions – need direct control over their AI systems. He warned against the risks of dependency on external providers who could potentially restrict access.
Yoshua Bengio, a leading AI researcher, echoed these concerns, emphasizing the need for a more equitable global order. “If you see the opposite [of a trustworthy partner], then you want to be preparing Plan B. It’s a question of democracy and a kind of equitable world order in which no one country can use technology to dominate the others,” he said.
The United States, however, signaled a reluctance towards centralized global governance. Michael Kratsios, director of the White House Office of Science and Technology Policy, advocated for “sovereign AI capability,” encouraging countries to adopt U.S. Technology as a foundation for building their own independent AI infrastructure. “Real AI sovereignty means owning and using best-in-class technology for the benefit of your people,” he said. “It does not mean waiting to participate in an AI-enabled global market until you have tried and failed to build full self-sufficiency.”
This stance was accompanied by a strengthening of strategic alliances. India joined Pax Silica, a U.S.-led technology alliance focused on securing semiconductor supply chains and countering Chinese AI efforts, alongside Japan, South Korea, the U.K., and Israel. This move represents a significant improvement in U.S.-India relations following recent disagreements over India’s purchases of Russian oil.
A Pair of Safety Commitments
Beyond the declaration, the summit yielded the New Delhi Frontier AI Impact Commitments, a set of voluntary agreements endorsed by leading AI companies. These commitments focus on transparency regarding AI usage – with companies agreeing to publish anonymized insights into the impact of their systems on jobs and productivity – and inclusion, through enhanced testing and evaluation of AI systems across diverse languages and cultural contexts, particularly in the Global South.
While welcomed as a positive step, some observers felt the commitments did not adequately address the full range of safety concerns discussed at the summit. Mark Brakel, director of policy at the Future of Life Institute, noted that critical issues such as child safety, national security risks, and the potential for loss of control were not sufficiently addressed in the final outcome.
Some policymakers characterized the summit as a natural progression from previous gatherings, shifting the focus from governance to commerce. While the summit did not establish a robust coalition of middle powers to challenge the dominance of the U.S. And China, conversations around alternative approaches to AI development were prominent.
A Flood of Investments
The summit proved to be a significant success in attracting investment to India. Electronics Minister Ashwini Vaishnaw announced that over $200 billion in AI and deep tech investment is expected in the country over the next two years. Reliance Industries pledged $110 billion over seven years to build AI and data infrastructure, while Adani Group committed $100 billion to renewable-powered AI data centres by 2035. Larsen & Toubro announced a venture with Nvidia to build a large-scale AI factory in India.
American tech companies also made substantial commitments. Microsoft plans to invest $50 billion across the Global South by 2030, building on previous investments in India. Google announced a $30 million AI for Government challenge and a $30 million AI for Science fund, alongside a new climate technology centre in partnership with the Indian government. Blackstone led a $600 million equity investment in Indian AI cloud startup Neysa, and AMD expanded its partnership with TCS to deploy AI infrastructure capacity.
OpenAI agreed to be the first customer for TCS’s data centre unit, and Anthropic revealed that India had become its second-largest market, opening a new office in Bengaluru.
While India successfully positioned itself as a key destination for AI investment, questions remain regarding its ability – and that of other nations outside the U.S.-China bloc – to shape the future of AI beyond simply participating in its growth.
