1,178 frontier lab employees asked Washington to build tools to slow AI down, and both OpenAI and Anthropic endorsed within hours.

Pacing the Frontier went public July 28 with signatures from OpenAI, Anthropic, Google DeepMind, Meta, Microsoft, Mistral, and Thinking Machines, asking the U.S. government to lead an international effort on the technical and governance tools needed to deliberately pace automated AI research. Signatories include Jakub Pachocki, Mark Chen, John Schulman, Jared Kaplan, Chris Olah, Jack Clark, Jan Leike, and Boris Cherny, who created and leads Claude Code. It lands four days before the administration's own deadline for a frontier-model security framework, the first U.S. document to define a covered frontier model and describe pre-release government access.