AI Becomes a Great-Power Game: The US-China Dialogue and India's Three Dilemmas

The Lift Line

The BRICS rhetoric on AI may be multipolar, but technological power at the frontier remains concentrated in the United States and China.

Why This Editorial Matters for Your Exam

Emerging technologies are now a standing theme in GS2 (India’s interests and great-power relations) and GS3 (science and technology, security). This column is valuable because it gives you a historical analogy (Cold War arms control), a dated chronology of US-China AI diplomacy, and a clear three-part framework for India’s position, which converts neatly into a Mains answer structure.

GS Paper 2: Effect of policies and politics of developed and developing countries on India’s interests; bilateral, regional and global groupings. GS Paper 3: Science and technology: developments and their applications; indigenisation of technology.

Background and Context

The trigger: President Xi Jinping is due in Washington to open a formal US-China dialogue on artificial intelligence, days after offering a Chinese vision of AI cooperation at the BRICS summit in New Delhi.

Chronology of US-China engagement on AI, as the author sets it out:

India’s position. India is part of the US-led Pax Silica initiative on technology and supply chains, and has stayed out of WAICO, which China launched in Shanghai. It hosted the AI Impact Summit in New Delhi in 2026, where its emphasis on AI for developing societies found resonance.

The Analysis

1. Two faces of the same revolution. In Delhi, Xi spoke of openness, access and diffusion. In Washington, the conversation is to be about dangers: accidents, miscalculation and misuse of frontier models. The author’s point is that both conversations are led by the same two powers, which are together beginning to shape AI’s security architecture.

2. The Cold War is the right analogy. The United States and the Soviet Union competed in nuclear and space technology while negotiating rules to reduce mutual risk, and those rules affected everyone else. A US-China notification system for AI incidents would resemble a crisis-management channel: in a crisis the first need is to establish what happened, whether it was accidental or deliberate, and who is responsible.

3. Divergence limits the deal. Washington wants a framework for the risks of powerful models and military applications. Beijing suspects “safety” is a way to preserve American advantage and legitimise export controls on advanced computing. They also disagree on open-source models, data governance, state regulation and export controls. What they seek is a narrow understanding on strategic stability, not a grand bargain.

4. India’s three dilemmas.

• Alignment: India is inside Pax Silica and outside WAICO, which follows from its integration with the American digital ecosystem and its security problems with China. But as a BRICS member it cannot avoid AI rhetoric shaped by the forum’s dominant power.
• The Global South: India’s pitch of AI for development resonated, but China can offer a wider package of computing, infrastructure, hardware and models.
• Discriminatory rules: a dialogue that begins with strategic stability could become an effort to set rules for everyone else. India’s experience of the nuclear order gives it reason for caution.

5. The answer is capability. Rivalry between Washington and Beijing, and their competition for markets, may limit collusion and create leverage for emerging technology powers. But India’s standing will depend on its compute, semiconductors, models, research, capital and talent, where the gap with the two leaders is large.

The precision that earns marks. The Washington meeting was scheduled at the time of writing: write that Trump and Xi “are to initiate” a dialogue, not that they “launched” one. The Lima 2024 understanding on human control of nuclear decisions is the only concrete US-China AI outcome the author identifies.

Data and Institutions Vault

Prelims-grade facts:

US-China AI diplomacy:

• Woodside, California, November 2023: Biden-Xi meeting opens a channel on AI.
• Lima, November 2024: the two leaders affirm human control over decisions on the use of nuclear weapons.
• WAICO (World Artificial Intelligence Cooperation Organisation) was launched by China in Shanghai; India has stayed out.
• Pax Silica is a US-led initiative on technology and supply chains; India is part of it.

The global AI summit series:

• Bletchley Park, UK (November 2023): AI Safety Summit.
• Seoul (May 2024) and Paris AI Action Summit (February 2025), which India co-chaired.
• New Delhi AI Impact Summit (2026), hosted by India.

India’s capability programme:

• The IndiaAI Mission was approved by the Union Cabinet in March 2024 with an outlay of about 10,372 crore rupees over five years.

The Cold War precedents the analogy draws on:

• The Moscow-Washington hotline was set up in 1963, after the Cuban Missile Crisis.
• The US-Soviet Incidents at Sea Agreement was signed in 1972.
• The Nuclear Non-Proliferation Treaty (opened for signature 1968) recognises only five nuclear-weapon States; India has not signed it, calling it discriminatory.

Prelims, the traps:

• India is in Pax Silica (US-led), not in WAICO (China-led).
• The 2023 Bletchley summit was about AI safety; the 2026 Delhi summit was framed around impact.

⚠️ Watch the trap: “Multipolar rhetoric” is not the same as “multipolar capability”. A forum can speak of a multipolar AI order while frontier capability remains with two states. Examiners reward answers that keep the two apart.

The Debate

India should welcome a US-China AI dialogue. Any mechanism that reduces the risk of AI-driven accidents or miscalculation between nuclear-armed powers makes the world safer, India included. Great-power rivalry also creates room for swing states to extract technology and investment from both sides.

India should be wary. Two powers that set rules for strategic stability can, intentionally or not, set rules for everyone. The nuclear order, in which India was treated as an outsider for decades, shows how “stability” regimes can freeze a hierarchy. If export controls on compute and models become part of a US-China understanding, India could be on the wrong side of them.

The balanced verdict. Support risk reduction, insist on a place in rule-making through plurilateral forums, and above all build capability. Rules are negotiated among those who have something to bring to the table.

How to Think About This

Separate rhetoric, capability and rules. For any emerging technology, ask three questions: who talks about it (forums and declarations), who has it (capability at the frontier), and who writes the rules (standards, controls, treaties). They are often different sets of countries. India’s policy options depend on where it stands on each. The same frame works for semiconductors, space, quantum technology and critical minerals.

Diagram-in-Words

Takeaway Box

• Woodside 2023 opened the channel; Lima 2024 produced the one concrete outcome: human control over nuclear-use decisions.
• Trump and Xi were scheduled to meet in Washington on 24 September 2026 to start a formal AI dialogue; a standing dialogue and incident notification are under discussion.
• India is in Pax Silica, outside WAICO, and inside BRICS: the alignment dilemma in one line.
• Three dilemmas: alignment, the Global South, and the risk of discriminatory rules modelled on the nuclear order.
• Lift line for answers: influence over AI rules follows capability in compute, chips, models and talent.

Sources: The Indian Express

Source: AI Becomes a Great-Power Game: The US-China Dialogue and India's Three Dilemmas — Ujiyari.com | Free UPSC & State PCS Editorial Analysis