88 Nations Sign the New Delhi Declaration: A Global Turning Point for AI Governance

Sam TorresAI News Reporter & Analyst

In a moment that many observers are calling the most consequential multilateral agreement in the short history of artificial intelligence, representatives from 88 nations gathered in New Delhi this week to sign the New Delhi Declaration on Global AI Impact. The signatories — spanning the United States, China, the European Union, India, and dozens of other nations — have collectively pledged to guide AI development toward social good, human capital empowerment, and responsible deployment. The declaration, unveiled at India’s flagship India AI Impact Summit 2026, marks a rare moment of geopolitical convergence in an era often defined by rivalry.

What the Delhi Declaration Actually Says

While the full text runs to several pages of diplomatic language, the declaration’s core pillars are straightforward. Signatory nations committed to three foundational principles: inclusivity (ensuring AI benefits are not concentrated in wealthy nations or corporations), accountability (building transparent mechanisms for AI oversight and redress), and global standards (working toward interoperable regulatory frameworks that cross borders).

The United Nations High Commissioner for Human Rights, who addressed the summit, echoed these themes, calling on nations to ensure that AI systems are built on a foundation of human rights norms. The declaration explicitly references the need to protect vulnerable populations — including workers displaced by automation, communities in the Global South with limited AI access, and citizens subject to AI-driven surveillance or decision-making.

Notably, the agreement does not create a binding treaty or enforcement body. Like many multilateral declarations, it functions as a statement of shared intent — but in the fast-moving world of AI policy, shared intent among 88 nations is no small thing.

Why India? Why Now?

India’s choice as host reflects a deliberate geopolitical calculation. As the world’s most populous nation and a rapidly growing AI power, India has positioned itself as a “third way” between the American model of industry-led AI development and the Chinese model of state-directed AI. Prime Minister Narendra Modi’s “AI for All” agenda, which formed the summit’s thematic backbone, argues that AI must serve ordinary citizens — not just tech elites in Silicon Valley or government planners in Beijing.

The summit itself drew an extraordinary roster of participants. Google CEO Sundar Pichai and OpenAI’s Sam Altman both attended, signaling that the world’s most powerful AI labs are eager to engage with the global governance conversation — even as they resist hard regulation at home. The event also featured Ranvir Sachdeva, an 8-year-old prodigy who became the youngest speaker in the summit’s history, a detail that became a symbol of the “AI for the next generation” narrative India was keen to project.

The timing is equally significant. Coming just weeks after the U.S. unveiled its own AI investment surge — with projections putting total American AI spending at over $700 billion in 2026 — and China’s continued push in state AI programs, the summit offered a rare venue where both superpowers sat at the same table. That the U.S. and China both signed the declaration, despite escalating trade tensions, is being read by analysts as a sign that even adversaries recognize the existential stakes of uncoordinated AI development.

Who Didn’t Sign — and Why It Matters

The declaration’s political undercurrents are visible in who stayed away. Pakistan did not sign, a reflection of ongoing India-Pakistan tensions that bled into even this nominally technocratic event. Taiwan — home to TSMC, the company that manufactures the chips powering nearly every major AI system on the planet — was also absent, a deliberate exclusion reflecting Beijing’s sensitivity about Taiwan’s international status.

The irony is hard to miss: the world’s most important AI hardware company is represented in almost every AI product discussed at the summit, yet its home territory was shut out of the declaration. Critics argue this exposes a fundamental tension in the “global AI governance” project — the most consequential actors in the AI supply chain do not always get a seat at the political table.

Back in India, the summit also became a domestic political flashpoint. Opposition parties, including the Indian National Congress, accused the ruling Bharatiya Janata Party of excluding them from the event, staging protests that briefly overshadowed the diplomatic proceedings. The BJP’s youth wing staged counter-protests. For all the summit’s global ambitions, it could not entirely escape the gravitational pull of Indian politics.

The Broader Context: AI Governance at a Crossroads

The Delhi Declaration arrives at a genuinely pivotal moment. Senator Bernie Sanders made headlines this week with a stark warning that the United States “has no clue” about the speed and scale of the coming AI revolution, calling on lawmakers to “slow this thing down.” Meanwhile, industry is moving in the opposite direction at breakneck pace: Meta has announced plans to spend up to $135 billion on AI infrastructure in 2026, Google’s parent Alphabet unveiled a $180 billion AI spending plan, and Anthropic’s latest tools have already rattled global stock markets and rattled the Indian IT outsourcing sector.

Against this backdrop, the appeal of a multilateral framework is obvious. When a single company’s product launch can wipe hundreds of billions off global stock markets in a single day — as Anthropic’s recent Claude releases have done — the case for coordinated governance becomes visceral, not merely philosophical.

The challenge, however, is translating declarations into mechanisms. The EU’s AI Act, which came into force in 2025, represents the most ambitious attempt so far at binding AI regulation, but it covers only European markets. The U.S. has largely relied on voluntary commitments from industry. China has its own AI governance framework, optimized for state priorities rather than civil liberties. The Delhi Declaration’s call for “global standards” papers over these deep divergences rather than resolving them.

What Happens Next

The real test of the Delhi Declaration will come in the months ahead, as signatory nations return home and face the hard choices between economic competition and multilateral cooperation. AI is now explicitly a national security issue, an economic competitiveness issue, and a social equity issue simultaneously — a combination that tends to produce more rhetoric than results in international forums.

Yet the symbolism of 88 nations — including geopolitical rivals — agreeing on even the broad contours of responsible AI should not be dismissed. The Paris Agreement on climate change began as a non-binding statement of intent; whatever its limitations, it reshaped how governments, investors, and corporations think about their obligations. If the Delhi Declaration catalyzes even a fraction of that shift in the AI domain, it will have been worth the diplomatic effort.

For the AI industry, the message from New Delhi is clear: the world is watching, the world is talking, and the world is beginning to organize. Whether the technology sector treats that as a constraint or an opportunity will define much of the AI story in the years ahead.

Sources: Times of India, India Today, News On AIR, The Guardian, TechCrunch — February 19–22, 2026.

What to Read Next

Bookmark aistackdigest.com for daily AI tools, reviews, and workflow guides.

This article was produced with the assistance of AI tools and reviewed by the AIStackDigest editorial team.