Over 1,100 AI Workers From OpenAI, Anthropic, Google and Meta Demand US Back Global Brakes on Frontier AI

More than 1,100 employees from the world's leading artificial intelligence labs signed an open letter on Tuesday asking the US government to back international tools that can deliberately pace frontier AI development. The signatories come from OpenAI, Anthropic, Meta, Google, Thinking Machines, Microsoft, and Mistral, according to The Verge. Bloomberg reports that the list includes Anthropic's CEO and other senior figures. The initiative, called Pacing the Frontier, is supported by nonprofits Guidelight AI Standards and Encode AI, Reuters notes.

The letter doesn't call for a blanket pause. Instead, it asks Washington to support a mechanism for coordinated global governance that can apply brakes if AI capabilities start accelerating beyond human control. Engadget reports that the workers want the government to help deliberately pace the development of transformative technology. The scale of the signatures, crossing multiple competing labs, marks one of the broadest internal calls for AI governance from the industry's own workforce.

The petition zeroes in on a specific near-term risk: AI systems that can automate their own research and improvement. The letter states plainly that the world's leading AI companies believe they could be close to automating AI research, according to The Verge. Businessinsider reports that the letter highlights concerns that frontier AI models may soon be able to improve themselves, which could drastically change the future of AI.

This isn't a vague existential worry. It's a concrete technical milestone that multiple labs are racing toward. If AI can recursively improve itself, capability development could rapidly accelerate in ways that are hard to predict or govern. The letter frames this as a real risk that demands proactive tools, not reactive panic. Reuters confirms that the statement comes as AI companies pursue increasingly powerful systems, raising concerns that developments could outpace existing safety, governance, and oversight mechanisms.

The signatories aren't asking Congress to pass a sweeping new law. They're asking the US government to support an international effort to develop governance tools. Reuters describes this as an international effort to develop tools for managing the pace of advanced AI development. Bloomberg adds that the petition calls on the US government to support a mechanism that would help deliberately pace AI development.

The distinction matters. The letter targets executive action and diplomatic engagement, not legislation. It asks the US to participate in building the infrastructure for coordinated international oversight. The Verge frames it as a slowdown of sorts for frontier AI development, or at least a speed-up of global coordinated governance efforts. The workers want the government to have options ready before a crisis, not scramble after one starts.

This letter represents a notable evolution in how AI workers engage with government. Earlier industry petitions, like the 2023 pause letter, were often led by outsiders or academics. This one comes from inside the labs, with senior officials at companies like Anthropic and OpenAI signing on, as Businessinsider reports. That internal origin changes the political calculus. These aren't critics throwing stones from outside; they're builders asking for guardrails on their own work.

Bloomberg notes that the signatories include the CEO of Anthropic, giving the letter institutional weight. The Verge reports that employees from Microsoft and Mistral joined as well, broadening the coalition beyond the usual suspects. The letter also arrives as the Trump administration's AI policy is taking shape, making the timing a pointed signal to policymakers that the industry's own workers want external oversight, not just self-regulation.

The letter exposes a tension that's been simmering inside AI labs: employees who build these systems are increasingly vocal about their risks, even as their employers race forward. The signatories work at companies that are actively competing to build the most powerful models. Yet over 1,100 of them are asking for a mechanism that could slow that very race.

This isn't the first time AI workers have gone public with safety concerns, but it's the largest coordinated statement from across the industry. Engadget reports that the petition includes workers from OpenAI, Anthropic, Google, Meta and more. The cross-company nature of the letter makes it harder to dismiss as a single company's internal dissent. It's a signal that safety concerns aren't isolated to one lab's culture but are widespread across the frontier AI workforce.

The letter's focus on international tools reflects a recognition that AI governance can't stop at any single country's borders. The workers want the US to engage with global partners to build shared mechanisms for pacing development. Reuters reports that the initiative is supported by nonprofits Guidelight AI Standards and Encode AI, which suggests there's an existing infrastructure of safety advocates ready to help operationalize these tools.

The challenge is that international AI governance remains mostly aspirational. The UK's AI Safety Summit, the EU's AI Act, and various UN efforts have produced frameworks but not the kind of real-time intervention tools the letter envisions. The workers are essentially asking the US to take a leadership role in turning those frameworks into actionable mechanisms. Whether the current administration or the next one will engage seriously with that request is an open question.