The White House is working with Anthropic on a framework that would let the U.S. government step in if a new AI model appears to pose serious national security or economic risks. The talks suggest Washington is moving away from ad hoc panic mode and toward a more formal approval-and-review process for powerful AI systems, […]