AI Chiefs Call for Slowdown — But Washington and Beijing Disagree

The leaders of the world’s most powerful AI labs have reached a rare consensus: it’s time to slow down. In an open letter, Anthropic CEO Dario Amodei called to “pace the frontier” of AI development, a proposal that quickly drew public support from OpenAI's Sam Altman, Google DeepMind's Demis Hassabis, and Elon Musk, according to reports from both CNBC and The Verge.

This public alignment among fierce competitors is, on its own, a significant market signal.

The consensus, however, is fragile and confined to the industry's inner circle. Outside of Silicon Valley, the reception is decidedly cooler, exposing the immense political and geopolitical hurdles that make any such slowdown unlikely in practice.

A Rare Consensus in Silicon Valley

Amodei's proposal, detailed in a lengthy open letter, suggests a coordinated effort to moderate the pace at which companies develop increasingly powerful AI models. The Verge notes that support from competitors was swift, with OpenAI’s Sam Altman and Elon Musk publicly agreeing on X. CNBC adds that Demis Hassabis, who heads Alphabet’s Google DeepMind, also offered his tentative support.

This unified front from the very individuals driving the AI race is unprecedented. These are the executives whose companies are locked in a capital-intensive battle for talent, computing power, and market share. Their collective call for a pause suggests a genuine concern about the trajectory of the technology they are building.

Yet, the actions of these companies often conflict with their public statements. The call for a slowdown comes even as major labs continue shipping more powerful products. OpenAI, for instance, recently released its GPT-6 Astra model, demonstrating that the underlying competitive drive to build and release next-generation systems has not abated. Taken together, these reports indicate that while industry leaders may agree on the destination—safer AI—they have not agreed to stop driving.

Political and Geopolitical Roadblocks

The first major obstacle is political. According to The Verge, top Republican figures, including former President Donald Trump and House Speaker Mike Johnson, believe the industry is overreacting. Their skepticism suggests that any attempt to enact a federally mandated or regulated slowdown would face significant political opposition in Washington.

This creates a difficult environment for AI labs, which may be seeking regulatory guidance but are instead met with accusations of alarmism.

The second, and perhaps more significant, roadblock is geopolitical. In his own comments, Amodei acknowledged that China presents the “toughest dilemma” for his proposal, as reported by CNBC. A voluntary pause by Western companies like Anthropic, OpenAI, and Google would be a strategic gamble if Chinese AI labs continue to accelerate their own development. Such a scenario could allow China to close the gap or even surpass the U.S. in a technology with critical economic and national security implications.

The data points to a fundamental conflict. The very executives calling for caution are also those best positioned to understand the competitive threat. Their plea for a slowdown is simultaneously a confession that they are in a race they feel they cannot afford to lose, or even pause, without a global agreement that currently seems out of reach.