Researchers and employees at leading artificial intelligence companies are calling for a slowdown of AI development to prioritize safety and security [1].

This movement signals a growing rift between the technical staff building these systems and the corporate leadership driving their release. If the U.S. government adopts these requests, it could fundamentally alter the pace of innovation and the competitive landscape of the global tech industry.

More than 1,000 employees at major AI firms, including OpenAI and Anthropic, signed a petition on Tuesday urging the government to help pace the release of advanced models [2]. The signatories said safety and security measures are lagging behind the rapid advances of the technology [1, 3]. They said they fear potential risks if the development of these systems continues unchecked [3].

These calls for a pause coincide with meetings in Washington, D.C., involving OpenAI CEO Sam Altman and other industry leaders [1, 2]. While researchers push for a slowdown, Altman and other executives said they support pacing the rapid advancement of AI capabilities [2].

The tension highlights a critical debate within the industry about the balance between speed and caution. The researchers said the current trajectory of development creates vulnerabilities that cannot be patched after a model is already public [3].

Lawmakers are now facing pressure to decide whether to implement regulatory frameworks that would mandate slower release cycles. Such measures would likely require a new oversight mechanism to determine when a model is sufficiently safe for general deployment [2].

More than 1,000 employees at leading AI companies signed a petition calling for a slowdown.

The push for a government-mandated slowdown suggests that internal corporate safety protocols are viewed as insufficient by the people implementing them. By appealing to Washington, these researchers are attempting to move AI safety from a voluntary corporate policy to a legal requirement, which would shift the power dynamic from CEOs to regulators.