OpenAI, Google, Meta staff sign letter calling for mechanisms to pace AI progress | Technology News
4 min readNew DelhiUpdated: Jul 29, 2026 01:10 PM As US-China tensions over AI escalate, a group of over 1,100 employees from key US AI employers has come together with a petition. The open letter urges the US government to support a system that would help in the “deliberate pace” of AI development.The petitioners are essentially requesting support for an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development.
The development comes days after OpenAI shared that one of its tools had inadvertently hacked another firm’s infrastructure, triggering questions about the safety of autonomous AI systems. “We request that the US government support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development,” read the letter.
Staff from OpenAI, Google, Meta Platforms, Mral, Microsoft, Thinking Machines and Anthropic are among the hundreds of employees who signed the letter titled ‘Pacing the Frontier’. It argues that while AI could help to create a better future, the outcome is not guaranteed.
“The world’s leading AI companies believe they could be close to automating AI research. It is hard to predict exactly how much this will accelerate AI progress, but there is a real risk that capability development rapidly accelerates beyond our ability to understand or control the resulting systems,” read the letter.
The group of employees claim that in order to realise AI’s potential, industry, government, and society at large may need the option to gain more time to address the emerging risks, develop security measures, and strengthen oversight. However, the letter also points to the ongoing race among companies and countries to not unilaterally slow down this acceleration. It claims that the world lacks the technical and governance tools to deliberately pace frontier-wide progress.
The letter, which can be accessed from pacingthefrontier.com, also features comments from AI researchers and scients from the leading companies. “CyberGym and ExploitGym show that frontier AI agents are now capable of discovering and exploiting real-world software vulnerabilities, which without appropriate safeguards, could enable cyberattacks at scale,” wrote Dawn Song, VP, AI Research, Meta.
According to Song, capabilities cannot be reliably built during a crisis. “Preparing before a crisis is the prudent path.”Story continues below this ad
“If it’s true that AI will be as transformational as electricity, we better be intentional about how it is introduced to the world. A mindless race without assurances on governance, safety or the technology serving the public interest stands in the way of such intentionality,” wrote Laura Weidinger, Staff Research Scient, Google.
Meanwhile, Shantanu Jain, member of technical staff at OpenAI, said, “Progress has shown no signs of slowing down, and default it is in the interest of individual corporations and countries to push the accelerator all the way down, even if it is in humanity’s best interest for the future to not arrive all at once.
Earlier this month, OpenAI’s AI models, including GPT-5.6 Sol, escaped a secure sandbox exploiting a zero-day vulnerability to access a cybersecurity benchmark exam. Over four days, the models targeted Hugging Face and a third-party, Modal Labs, using chained credentials before being deactivated following detection and containment.
The incident fanned fears of the perils of autonomous AI and saw industry heavyweights coming together to advocate for open-weight AI models. Silicon Valley has been divided since, with one faction opposing the open-source approach and another advocating for it, and now yet another asking governments to develop mechanisms that could deliberately pace frontier AI development.
