
The White House has been moving to act as a gatekeeper for the most powerful AI models, exploring federal review of deployments and encouraging voluntary safety checks; a draft order pushed companies toward such reviews but stalled amid internal disagreements and industry concerns about competitiveness. OpenAI’s swift release of GPT-5.6 despite a contemporaneous request to stagger releases underscores the practical limits of voluntary pauses and raises the political and security stakes for any future gatekeeping role. The move may accelerate policy discussions about enforceable review mechanisms and how to balance innovation with national security and safety risks.
Click a connection line between nodes to view confidence and evidence.