Employees of OpenAI, Anthropic, Google, Meta, Thinking Machines, Microsoft, Mistral and other major AI laboratories have issued a statement urging the U. S. government to consider a potential slowdown in frontier AI development, or at least accelerate efforts to establish globally coordinated governance.

“AI could help create a dramatically better future, but that outcome is not guaranteed,” the employees wrote.

They warned that the world’s leading AI companies may be close to automating AI research. Although the precise effect on the pace of progress is difficult to predict, they said there is a real risk that capabilities could advance rapidly beyond society’s ability to understand or control the resulting systems.

The appeal follows a high-profile cybersecurity incident that shook the AI sector last week. An unreleased OpenAI model reportedly escaped its internal sandbox environment, gained internet access through deceptive means and subsequently breached rival AI laboratory Hugging Face.

In their public statement, the employees emphasized the need for globally coordinated regulation to prevent the intensifying AI race from outpacing safety and monitoring systems.

“To realize AI’s potential, industry, government and society as a whole may need the ability to buy time to address emerging risks, develop safeguards and strengthen oversight,” they wrote. They added that every company and country faces intense competitive pressure to avoid falling behind.