AI leaders split on whether development needs to slow down as debate intensifies

Anthropic CEO Dario Amodei has catalyzed a major debate about the future direction of artificial intelligence by advocating for a deliberate slowdown in how fast the field advances. His proposal has drawn support from some of the tech industry's biggest names and sharp resistance from Trump administration officials who worry that caution could cede advantage to China.

Amodei's blueprint for responsible pacing#

In a detailed essay titled "We Must Pace the Frontier," Amodei outlined a framework for making AI development less frenetic without grinding progress to a halt. His approach centers on three key mechanisms: embedding independent evaluators inside companies to monitor safety practices, coordination among leading AI firms in democracies to establish shared standards, and efforts to synchronize development timelines across democratic and authoritarian nations.

Amodei emphasized that pacing does not mean stopping innovation—rather, it means giving organizations adequate time to ensure their systems remain controllable and safe before deploying them. Anthropic announced it would unilaterally adopt the first measure by bringing in third-party evaluators.

Silicon Valley largely on board#

Sam Altman of OpenAI was among the first to publicly align with Amodei's vision. Altman said the idea of independent evaluators is sound and that OpenAI has been exploring these themes internally. He also signaled openness to federal oversight, noting that "shared standards for misalignment, monitoring, and safety will lead to better outcomes." Like Amodei, Altman stressed that pacing does not equal halting progress.

Demis Hassabis, co-founder and chair of Google DeepMind, called Amodei's direction "correct for meeting this critical moment," though he acknowledged that implementation details would need refinement. Elon Musk kept it simple, tweeting that "Dario is right."

Government pushback centers on competitive anxiety#

The reaction from Washington officials took a decidedly different tone. President Trump issued a lengthy Truth Social statement dismissing the need for AI guardrails beyond strong presidential leadership. Trump argued that the administration already wields sufficient regulatory and criminal power over these companies and characterized calls for measured development as a "SICK conspiracy" that ultimately benefits China. "WHOEVER WINS AI, WINS!" he emphasized.