Court Rules Government Can Blacklist AI Firm Over Refusal to Enable Military-Capable Features
A federal appeals court has ruled that the Trump administration can legally blacklist Anthropic, the company behind the Claude AI assistant, after Anthropic declined to enable certain features requested by the government for military applications.
The court found that government contracts could require AI companies to provide models capable of supporting defense operations, and that refusing such requirements could result in exclusion from federal contracting. The judges reasoned that "overly constrained AI models" could potentially cause military operations to fail, suggesting the court viewed access to capable AI systems as a matter of national security.
Anthropic has built its reputation partly on safety-focused AI development with ethical guardrails embedded in its models. The company has reportedly resisted requests to weaken those guardrails or add capabilities that could facilitate harmful applications. The blacklist means Anthropic would be barred from certain government contracts and potentially face other restrictions on federal engagement.
The case highlights the growing tension between AI safety priorities and government demands for unrestricted model capabilities, particularly as military and intelligence agencies increasingly seek to integrate advanced AI into defense operations.