Compare the Western (G7/EU) and Chinese approaches to AI governance. How should middle powers like India navigate this bifurcation?

Global AI governance is splitting into two architectures — the EU's binding, rights-based AI Act and China's Shanghai-headquartered World AI Cooperation Organisation (WAICO), launched in July 2026 [3][4]. For middle powers, the task is not to pick a bloc but to widen the multilateral middle.

Western (G7/EU) approach: rules-first

• Regulation as the instrument: the EU AI Act (Regulation 2024/1689) tiers systems by risk, is legally binding and has extraterritorial reach [4]; G7-led codes of conduct add voluntary industry commitments.
• Anchor values: safety, transparency, fundamental rights and accountability of developers.
• Limitation: high compliance costs and standards written largely without developing-country input — precisely the exclusion the UN Secretary-General warned against, that AI "cannot be governed by a handful of countries or companies" [2].

Chinese approach: institution-first

• Builds bodies, not binding rules: WAICO, founded by 29 countries mostly from Asia, Africa and Latin America, follows the Global AI Governance Initiative (2023) and the AI Global Governance Action Plan of WAIC 2025 [3][6].
• Capacity-building as currency: training offers and joint AI application centres with ASEAN, the African Union, SCO and BRICS [3].
• Limitation: state-sovereignty-heavy, thin on enforceable safety obligations, and risks fragmenting the UN track.

How India should navigate

• Anchor in the UN: back the Global Dialogue on AI Governance, mandated by UNGA Resolution A/RES/79/325, as the legitimate universal forum [1].
• Lead, not follow: the India AI Impact Summit 2026 at Bharat Mandapam, with 100+ countries, shifted the agenda from "AI safety" to "AI for development" [5] — India's distinct third path.
• Issue-based alignment: partner the West on compute, chips and safety research; work with the Global South on affordable models and skilling, without treaty entanglement.
• Strategic autonomy through capability: scale the IndiaAI Mission's compute and open-model stack, since rule-making power follows technological capacity [5].

India's advantage is credibility with both camps. By coupling domestic capability with a development-first, UN-anchored agenda, it can convert bifurcation into bargaining power and make AI governance genuinely inclusive.