State Department Drops Anthropic for OpenAI as Trump Administration Cancels AI Contracts

The U.S. State Department has migrated its internal chatbot from Anthropic's Claude Sonnet 4.5 to OpenAI's GPT-4.1, following a Trump administration directive to cancel federal contracts with Anthropic. The switch, first reported by Nextgov/FCW through internal documents, marks a significant shift in government AI procurement. The move comes amid a broader DOD supply chain risk designation targeting Anthropic, which Microsoft has challenged by filing an amicus brief advocating for a temporary restraining order to block the designation. Microsoft's intervention signals that major tech companies view the government's approach to AI vendor restrictions as potentially harmful to competition and national security readiness. The company's brief argues for judicial intervention to prevent the DOD designation from taking effect. The Anthropic situation reflects the growing entanglement of AI policy with political considerations. Federal agencies that adopted Anthropic's models now face pressure to switch providers regardless of technical merit or integration costs. For the broader AI industry, the episode raises questions about vendor lock-in risks when government contracts can be redirected by executive action rather than competitive evaluation.