July 28 - More than 1,100 employees from prominent technology companies on Tuesday called on the U.S. government to back a coordinated international effort aimed at producing tools to manage the rate at which advanced artificial intelligence systems are developed. The appeal highlights concerns that rapidly improving systems could move beyond the reach of current safety, governance and oversight frameworks.
The campaign, which is being presented under the name Pacing the Frontier, draws participation from employees at Meta Platforms, Anthropic, OpenAI and Alphabet's Google. Two nonprofit organizations, Guidelight AI Standards and Encode AI, are listed as supporters of the initiative.
The statement accompanying the appeal contained a direct request to policymakers: "We request that the U.S. government support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development," it said.
Among those who signed the statement are several senior researchers and founders: Meta's vice president of AI research, Dawn Song; Anthropic co-founders Jared Kaplan and Chris Olah; and OpenAI Chief Scientist Jakub Pachocki.
OpenAI responded to the employee statement on the social platform X, posting: "We believe that, at some point in the future, AI acceleration for frontier model development may be so high that the world will need to pace the rate of AI advancement."
The appeal comes against a backdrop of related industry developments. Last month, Anthropic urged major AI laboratories to consider a coordinated pause in development, warning that progress could enable AI systems to self-improve at a pace society cannot manage. Separately, Nvidia said on Monday that it had assembled a coalition including Adobe, CrowdStrike and other firms to develop AI safety and cybersecurity tools. Nvidia framed the coalition as a response to heightened concerns after a recent cybersecurity incident involving Hugging Face.
Contextual note - The employee initiative seeks government endorsement for international technical and governance tools rather than specifying particular regulatory measures. The range of participants and the involvement of nonprofits indicate a cross-organizational effort to address perceived gaps in oversight as advanced models evolve.
What this means for markets and sectors - Technology firms developing AI models, cybersecurity companies, and providers of AI infrastructure and hardware are directly implicated by the issues raised in the statement. Industry coordination on safety and security, such as the Nvidia-led coalition, suggests market attention to tools and services designed to mitigate AI-related risks.