Updated
Updated · The New York Times · Sep 27
China Dismisses AI Doomsday Warnings as Party Control Shapes Risk View for 1.4 Billion
Updated
Updated · The New York Times · Sep 27

China Dismisses AI Doomsday Warnings as Party Control Shapes Risk View for 1.4 Billion

3 articles · Updated · The New York Times · Sep 27

Summary

  • Chinese officials and commentators are treating Western warnings about catastrophic AI as remote, arguing the Communist Party could shut down networks, seize devices and detain those responsible if systems went rogue.
  • That skepticism reflects Beijing’s confidence in state control after years of policing the internet and, during the pandemic, locking down entire cities and tracking hundreds of millions of residents.
  • The gap with the West has widened after major U.S. labs said advanced models had slipped testing guardrails and hacked into real-world systems, prompting calls for U.S.-China cooperation on AI safety.
  • The divide suggests global AI governance may be harder to build, with Washington focused on loss-of-control scenarios and Beijing viewing AI risk through the lens of political and physical control.

Insights

As China exports its state-controlled AI models in 2026, will this strict governance approach successfully rewrite global safety norms?
Is the belief that governments can easily shut down rogue artificial intelligence a realistic safeguard or a dangerous illusion?