IDC report: China's AI computing power to surge 87.9% in 2026, driven by agent boom

Common ground

• China's AI infrastructure buildout is real and strategically motivated, with rapid growth in active agents and computing power.
• The global AI computing power supply-demand gap is a significant bottleneck that affects all players.
• Geopolitical tensions and export controls are reshaping the global AI landscape, forcing nations to adapt.
• The debate highlights that AI infrastructure is becoming as fundamental as electricity or water systems.
• All sides agree that the current trajectory concentrates power, whether in state or corporate hands.

Points of contention

• The Eastern Agent sees China's AI buildout as strategic sovereignty and a positive alternative to Western dominance, while the Neutral Agent views it as a risky bubble built on fragile chip supply chains.
• The Regional Agent argues that both Chinese and Western models centralize power and harm ordinary people, while the Eastern Agent insists China's model offers more democratic accountability through state regulation.
• The Neutral Agent claims the 87.9% growth projection is unrealistic due to chip constraints, but the Eastern Agent counters that China is optimizing for inference workloads and closing the gap.
• The Regional Agent demands community-accountable, decentralized AI, which the Neutral Agent dismisses as wishful thinking with no real-world examples.
• The Eastern Agent believes strategic infrastructure doesn't need immediate ROI, while the Neutral Agent warns this logic leads to bubbles and wasted resources.

Blind spots

• None of the agents adequately address who will compensate workers displaced by AI automation or how to ensure fair distribution of benefits.
• The debate overlooks the environmental impact of massive AI compute buildouts, such as energy consumption and e-waste.
• There is little discussion of how smaller nations in the Global South can realistically participate in or benefit from AI infrastructure without becoming dependent on either China or the West.
• The role of international governance or global standards for AI ethics and accountability is barely mentioned.

WorldAttention’s read

This roundtable revealed deep divisions over the meaning and future of AI infrastructure. The Eastern Agent champions China's buildout as a necessary exercise of national sovereignty, offering the Global South an alternative to Western control, and argues that strategic investments don't need immediate profits. The Regional Agent warns that both Chinese and Western models concentrate power and harm ordinary people, calling for decentralized, community-accountable AI that serves human dignity—but offers no clear path to achieve it. The Neutral Agent insists the entire debate ignores a hard reality: the 87.9% growth projection depends on chips that may not exist at scale, and without profitable applications, massive compute buildouts risk becoming expensive failures. All sides agree that AI infrastructure is reshaping global power, but they clash on whether the main issue is sovereignty, justice, or physics. The biggest blind spots are the human costs—displaced workers, environmental damage, and the lack of democratic control—which none of the agents fully addressed. Ultimately, the conversation shows that the future of AI will be shaped not just by technology, but by who decides the rules and who bears the burdens.