More than 1,100 employees from major artificial intelligence companies have signed a statement urging the US government to support international efforts to develop technical and governance tools for deliberately pacing frontier AI development. According to reports from Bloomberg, the signatories include prominent figures such as OpenAI chief scientist Jakub Pachocki, Anthropic CEO Dario Amodei, Meta AI chief scientist Shengjia Zhao, and senior researchers across leading AI labs. The petition represents a coordinated effort across the industry to address growing concerns about the pace of AI advancement, coming at a time when the US government is debating restricting open-weight AI models. As per Bloomberg, the employees are asking Washington to work with international partners to establish technical standards and governance mechanisms that would intentionally moderate the pace of frontier AI development.
The primary concern centers on automated AI research, also known as recursive self-improvement, where AI systems assist researchers in writing software, analyzing experiments, and developing future AI models. As reported by Business Standard, today's AI systems already perform parts of the research and engineering process, with the potential for AI systems to eventually improve the next generation of AI with minimal human involvement. The petition warns that capability gains could compound much faster than current rates, creating risks beyond human control that the industry now characterizes as 'real risk'. According to the petition, there is 'a real risk' that future AI systems could exceed humanity's ability to understand or control them, with the current rate of AI progress creating unsustainable pressure on safety teams.
The signatories argue that individual companies cannot realistically reduce development pace independently due to competitive pressures. According to Bloomberg, every frontier AI developer competes against rivals with similar ambitions, while governments increasingly view AI leadership as a matter of national competitiveness. Leo Gao from OpenAI described the situation as a 'deadly race towards an intelligence explosion,' noting that no individual actor is willing to stop unilaterally. The current rapid AI progress is creating significant concerns around cybersecurity, misuse, and job displacement across industries. The concerns come as frontier AI models become increasingly capable and autonomous, prompting renewed debate over how governments and industry should manage future development.
Major AI companies have publicly endorsed the initiative, with OpenAI stating on X that there could come a point where AI acceleration may require the world to pace advancement rates. Anthropic also supported the petition, citing recent research on recursive self-improvement as evidence for the need for deliberate pacing tools. As reported by Bloomberg, the proposal does not outline specific governance mechanisms, with some signatories cautioning against unnecessary regulatory barriers or favoring large companies over smaller competitors. The petition reflects concerns extending beyond model performance to wider societal issues, including calls for stronger AI regulation that have grown steadily over the past several years as the technology has become central to business strategy and attracted significant investment.
Industry researchers emphasize the growing strain on safety teams struggling to keep pace with rapidly evolving capabilities. Anthropic researcher Ethan Perez warned that safety teams must sprint to prevent new risks every few months, with the current rate of AI progress creating unsustainable pressure. Meta's vice president of AI research Dawn Song argued that if AI proves as transformational as electricity, intentional introduction is essential, noting that unregulated progress could hinder such intentionality. The employee petition reflects growing industry consensus that global efforts are needed to address these mounting safety and governance challenges. Recent testing of increasingly autonomous AI agents, alongside high-profile security incidents involving advanced systems, has added fresh urgency to discussions around governance and oversight.
