1,100 Tech Workers Urge Government Review of AI Pacing

A significant employee mobilization has emerged under the banner 'Pacing the Frontier'. More than 1,100 tech workers from OpenAI, Google DeepMind, Anthropic, and Meta have signed an open letter calling on the US government to intervene and slow down advanced pretraining runs.

The petition demands a voluntary framework for pre-release review of models exceeding 10^26 FLOPs in compute scale. Engineers tracking alignment policy updates can organize their task milestones with [Simple Todos](/tools/simple-todos/).

Grassroots Pressure for AI Development Guardrails

Employees express concerns that the commercial race for artificial general intelligence is forcing labs to deprioritize safety alignment. The letter argues that post-launch patches are insufficient to contain agentic capabilities.

Rising Skepticism of Self-Regulation Claims

The movement represents a shift in employee activism, with researchers seeking external federal oversight over their own firms. The organizers plan to submit the petition to the Senate Commerce Committee next month.