.svg?width=1400)
In this analysis
The AI safety debate has acquired the most dangerous phrase in politics: 'we can't do this because China won't.' Beijing now mirrors the argument by portraying Western slowdown proposals as a Cold War tactic. Both sides can use the other as justification for permanent acceleration.
What happened
Reuters reported that Trump dismissed calls to slow AI, stressing US leadership over China, while China's state-backed Global Times attacked slowdown proposals as an attempt to preserve Western technological dominance.
The argument
Strategic competition is real, but it can become intellectually lazy. Nuclear rivals eventually built arms-control mechanisms not because they trusted one another but because uncontrolled competition was dangerous. Frontier AI may require the same logic.
The strongest counterargument
AI is harder to verify than nuclear arsenals, commercial actors move quickly and any pause could be exploited by competitors. Effective coordination may be technically and politically unrealistic.
What happens next
Mid-September US-China AI safety talks will indicate whether both governments can separate genuine risk reduction from industrial policy and export-control bargaining.
Editorial note
This article separates sourced reporting from analysis. The factual record is linked below; judgments and conclusions are State & Power's editorial analysis.