The White House announced that the Trump administration will permit selected U.S. firms to test Anthropic's new Mythos model, easing a months‑long standoff between the AI lab and federal regulators. The move grants limited API access while keeping the broader deployment under review, signalling a tentative compromise after earlier threats to block the technology.

Industry observers note the decision reflects Washington's piecemeal approach to AI oversight, where ad hoc permissions replace a coherent framework. Anthropic, backed by major investors, had halted certain collaborations pending clarity on export‑control rules. By allowing a narrow slice of the market to experiment, the administration hopes to gather data without granting full‑scale commercial launch.

Investors should watch how the restricted access impacts Anthropic's revenue pipeline and whether the limited rollout triggers broader policy shifts. The arrangement underscores the tension between rapid AI innovation and a regulatory environment still defining its boundaries.